Western University Western University

Scholarship@Western Scholarship@Western

Electronic Thesis and Dissertation Repository

10-28-2019 2:00 PM

Engineering self-assembling proteins to produce a safe and Engineering self-assembling proteins to produce a safe and

effective vaccine for Porcine Reproductive and Respiratory effective vaccine for Porcine Reproductive and Respiratory

Syndrome Syndrome

Ondre H. Harper, The University of Western Ontario

Supervisor: Menassa, Rima, Agriculture and Agri-Food Canada

Co-Supervisor: Hill, Kathleen, The University of Western Ontario

A thesis submitted in partial fulfillment of the requirements for the Master of Science degree in

Biology

© Ondre H. Harper 2019

Follow this and additional works at: https://ir.lib.uwo.ca/etd

Part of the Biochemistry Commons, and the Biology Commons

Recommended Citation Recommended Citation Harper, Ondre H., "Engineering self-assembling proteins to produce a safe and effective vaccine for Porcine Reproductive and Respiratory Syndrome" (2019). Electronic Thesis and Dissertation Repository. 6670. https://ir.lib.uwo.ca/etd/6670

This Dissertation/Thesis is brought to you for free and open access by Scholarship@Western. It has been accepted for inclusion in Electronic Thesis and Dissertation Repository by an authorized administrator of Scholarship@Western. For more information, please contact wlswadmin@uwo.ca.

ii

Abstract

Commercially available vaccines for porcine reproductive and respiratory syndrome virus

(PRRSV) provide some control over the virus but none are ideal since they either are not

completely safe for use, lack efficacy in promoting long-lasting immunity or provide no

protection from heterologous PRRSV strains. Innovative approaches to designing vaccines

are being pursued to overcome these drawbacks. One example is the use of nanoparticles to

present a dense array of antigenic epitopes to the immune system which can effectively

stimulate antibody producing cells (B cells) and T cells, resulting in long-lasting immunity.

Here, I genetically fused antigenic epitopes from PRRSV to multiple self-assembling protein

nanoparticles and assessed their ability to be produced recombinantly in E. coli. The most

successful candidate was purified to homogeneity and demonstrated via multiple biochemical

and biophysical techniques to assemble like the native nanoparticle. Immunological testing

will reveal the candidate’s efficacy as a vaccine against PRRSV.

Keywords

porcine reproductive and respiratory syndrome, subunit vaccine, protein nanoparticles,

Brucella lumazine synthase

iii

Summary for Lay Audience

The goal of my project is to produce a safe and effective vaccine for porcine reproductive

and respiratory syndrome virus (PRRSV), a devastating disease in pigs that accounts for over

$600 M in losses per year in the US. Commercially available vaccines for PRRSV provide

some control over the virus but none are ideal since some contain damaged but living viruses

that in time may cause disease in the pigs, some contain killed viruses or pieces of the virus

that lack efficacy in promoting protection against the native virus both short-term and long-

term, and some provide no protection against multiple strains of the virus. Vaccines

containing whole viruses may be less safe but are generally more effective than vaccines

containing free pieces of the virus. My approach to designing a new PRRSV vaccine that is

safe and effective involves presenting specific protein sequences from the PRRS virus to the

pig’s immune system in a virus-like fashion, without the virus. I will be using stable protein

nanostructures as carriers for the chosen peptides from PRRSV; these nanostructures can

effectively mimic the architecture of the virus while having no potential for replication. In

my work, I genetically fused peptides from PRRSV that are known to stimulate the pig

immune system to multiple protein nanostructures and determined their ability to be

produced in and purified from a bacterial expression host. The most promising nanostructure

was purified and studied using techniques that allow us to determine whether the attachment

of the chosen PRRSV peptide to the nanostructure affected its assembly and stability. I

demonstrated that I can successfully attach an immunoreactive peptide from PRRSV on to a

very stable nanostructure and this product has the potential of being both safe as it cannot

cause disease, and effective as it may stimulate a strong protective immune response against

PRRSV.

iv

Acknowledgments

I would like to thank my amazing supervisors, Dr. Christopher Garnham and Dr. Rima

Menassa, for giving me the opportunity to learn from them over these years and for giving

me constant guidance and encouragement. They have improved my abilities as a researcher

by challenging me to think more critically, write concisely and convey my thoughts

effectively; and for that I am truly grateful.

Many thanks to my co-supervisor, Dr. Kathleen Hill, and my committee member, Dr. Robert

Cumming, for their valued comments, suggestions and perspective regarding my work. For

their assistance in the lab, I would like to thank Dr. Patrick Telmer and Shane Butler.

I would like to thank Dr. Richard Gardiner for training on the electron microscope and Lee-

Ann Briere for training on the circular dichroism spectropolarimeter and assisting in the

collection and interpretation of my ultracentrifugation data.

Finally, to my family and friends, thank you for the immense love and support throughout

this milestone in my life.

v

Table of Contents

Abstract ............................................................................................................................... ii

Summary for Lay Audience ............................................................................................... iii

Acknowledgments.............................................................................................................. iv

Table of Contents ................................................................................................................ v

List of Tables ................................................................................................................... viii

List of Figures .................................................................................................................... ix

List of Abbreviations ........................................................................................................ xii

Chapter 1 ............................................................................................................................. 1

1 Introduction .................................................................................................................... 1

1.1 Porcine reproductive and respiratory syndrome virus (PRRSV) ............................ 2

1.1.1 The two major envelope proteins of PRRSV .............................................. 4

1.2 Vaccines and vaccination approaches ..................................................................... 7

1.2.1 PRRSV vaccines ......................................................................................... 8

1.3 Virus-like particles .................................................................................................. 8

1.4 Platforms of interest .............................................................................................. 15

1.4.1 Brucella lumazine synthase ...................................................................... 15

1.4.2 Aquifex aeolicus lumazine synthase.......................................................... 15

1.4.3 Ferritin....................................................................................................... 16

1.4.4 Small heat shock protein ........................................................................... 17

1.5 Rationale and goal................................................................................................. 17

1.6 Objectives ............................................................................................................. 17

Chapter 2 ........................................................................................................................... 19

2 Experimental Procedures ............................................................................................. 19

2.1 Construction of chimeras ...................................................................................... 19

vi

2.2 Expression and purification .................................................................................. 22

2.2.1 Expression and lysis .................................................................................. 22

2.2.2 Immobilized metal affinity chromatography ............................................ 22

2.2.3 Anion-exchange chromatography ............................................................. 23

2.2.4 Size-exclusion chromatography ................................................................ 23

2.3 Characterization .................................................................................................... 24

2.3.1 Transmission electron microscopy ........................................................... 24

2.3.2 Circular dichroism spectroscopy ............................................................... 24

2.3.3 Sedimentation velocity.............................................................................. 25

Chapter 3 ........................................................................................................................... 26

3 Results .......................................................................................................................... 26

3.1 The GP5-antigen chosen is conserved and immunogenic .................................... 26

3.2 Nanoparticles fused with the GP5-antigen are produced but are insoluble .......... 29

3.3 Nanoparticles are enriched and partially purified by immobilized metal-affinity

chromatography .................................................................................................... 32

3.4 The new M-GP5-BLS construct is an improvement on GP5-BLS ....................... 35

3.5 Size-exclusion chromatography indicates multimeric assembly of BLS and M-

GP5-BLS ............................................................................................................... 40

3.6 Sedimentation velocity analysis indicates that M-GP5-BLS is elongated while

BLS is globular ..................................................................................................... 42

3.7 Secondary structure and thermal stability are conserved in the M-GP5-BLS

chimera .................................................................................................................. 45

3.8 Transmission electron microscopy shows pentameric assembly of both BLS and

M-GP5-BLS .......................................................................................................... 49

Chapter 4 ........................................................................................................................... 51

4 Discussion .................................................................................................................... 51

4.1 Selection of candidate PRRSV vaccine ................................................................ 51

4.2 Structural characterization of BLS and M-GP5-BLS ........................................... 55

vii

4.3 Conclusion and future directions .......................................................................... 59

References ......................................................................................................................... 61

Appendices ........................................................................................................................ 73

Curriculum Vitae .............................................................................................................. 75

viii

List of Tables

Table 1. Output summary from CDPro showing percentages of secondary structure assigned

to each protein by three different programs (CONTINLL, CDSSTR and SELCON3) using

the CDPro protein reference set SMP56. ................................................................................ 48

ix

List of Figures

Figure 1. Schematic showing the structure of the PRRS virion.. ............................................. 3

Figure 2. Schematic diagram showing the predicted topology of the proteins M and GP5 of

PRRSV VR-2332 within the virus envelope. ........................................................................... 6

Figure 3. Summary of the different types of vaccines highlighting key drawbacks of each .. 12

Figure 4. Surface representation of the candidate antigen carriers dispayed in PyMOL ....... 14

Figure 5. Surface images of the Brucella lumazine synthase (BLS) fused to the extended

representation of the GP5 antigen, displayed in PyMOL ....................................................... 18

Figure 6. Schematic of the designed GP5-fusions .................................................................. 21

Figure 7. Multiple sequence alignment of the GP5 ectodomain of the PRRS virus from the

North American genotype and the European genotype .......................................................... 28

Figure 8. Expression and solubility of the wild-type or GP5 engineered protein nanoparticles

of interest ................................................................................................................................ 30

Figure 9. Expression and solubilization of engineered nanoparticles with urea..................... 31

Figure 10. Immobilized metal affinity chromatography ......................................................... 33

Figure 11. Refolding of engineered nanoparticles .................................................................. 34

Figure 12. Introduction to M-GP5-BLS construct .................................................................. 37

Figure 13. SDS-PAGE gel analysis of the expression, solubility, nickel purification and

refolding of M-GP5-BLS and GP5-M-BLS. .......................................................................... 39

Figure 14. Size-exclusion chromatography. ........................................................................... 41

Figure 15 Sedimentation coefficient distribution of BLS and M-GP5-BLS. ......................... 44

Figure 16. Circular dichroism (CD) spectroscopy of BLS and M-GP5-BLS ......................... 47

x

Figure 17. Negative stain electron microscopy analysis of BLS and M-GP5-BLS................ 50

xi

List of Appendices

Appendix 1. Amino acid sequences of all proteins produced recombinantly......................... 73

xii

List of Abbreviations

AaLS: Aquifex aeolicus lumazine synthase

AUC- analytical ultracentrifugation

BLS: Brucella lumazine synthase

CD: circular dichroism

EAV: equine arteritis virus

EU: European

GGS: glycine-glycine-serine

GP5: glycoprotein 5

HBcAg: hepatitis B core antigen

His: histidine

IBs: inclusion bodies

IMAC: immobilized metal affinity chromatography

IPTG: isopropyl β-D-1-thiogalactopyranoside

LDV: lactate-dehydrogenase elevating virus

MW: molecular weight

NA: North American

Ni- nickel

Ni-NTA: nickel-charged nitrilotriacetic acid

ORF: open reading frame

PRRSV: porcine reproductive and respiratory syndrome virus

RNA: ribonucleic acid

SDS-PAGE: sodium dodecyl sulfate polyacrylamide gel electrophoresis

sHSP: small heat shock protein

SV- sedimentation velocity

TEM: transmission electron microscopy

xiii

Tm: melting temperature

Tris-HCl: tris (hydroxymethyl) aminomethane hydrochloride

UV: ultraviolet

VLP: virus-like particle

1

Chapter 1

1 Introduction

Porcine reproductive and respiratory syndrome (PRRS) is a viral disease first observed in

the US in the 1980s where it was referred to as the mystery swine disease (Benfield et al.,

1992; Wensvoort et al., 1991) and it quickly spread throughout North America, being

reported in Canada in 1992 and in several European countries thereafter. Currently PRRS

is endemic in swine herds worldwide and spread between and within farms occurs most

commonly via the introduction of infected pigs to the farm, and direct contact with nasal

secretions, mammary secretions, semen, urine and feces of an infected pig. Although less

common today, due to the current biosecurity measures, transmission via artificial

insemination of sows with imported semen from an infected boar was a major cause of

introduction of the virus to herds in other countries (Nathues et al., 2016).

PRRS generally presents itself in two ways, as a reproductive disease in sows and a

respiratory disease in growing pigs. Sows infected with the virus have high rates of

abortions, stillbirths and even the delivery of mummified piglets. As for the piglets that

make it to term, they are generally weak and have low life expectancies (Christianson,

1992). In young pigs infected with the virus, fever, sneezing, pneumonia and stunting of

growth are common symptoms (Botner et al., 1997). Currently, PRRS is the most

economically significant swine disease affecting the US food industry. In 2012, annual

losses due to PRRS were estimated in the US to be $664 million, an increase from the

2005 estimate of $560 million per year (Holtkamp et al., 2013; Neumann et al., 2005). In

Canada PRRS was estimated to cost the swine industry $130 million per year (Johnson,

2012).

2

1.1 Porcine reproductive and respiratory syndrome virus (PRRSV)

PRRSV is a small enveloped RNA virus (50-65 nm) in the genus Arterivirus. Closely-

related arterivirus species include equine arteritis virus (EAV) and lactate-dehydrogenase

elevating virus (LDV) whose infection of horses and mice respectively, also leads to

respiratory failure and abortion (Snijder et al., 2013). Two genotypes of PRRSV exist, the

European (genotype 1; EU) and the North American (genotype 2; NA) and they are

genetically and antigenically distinct. The prototypic European Lelystad virus and North

American VR-2332 isolates, possess only about 60 % nucleotide identity and this

divergence is thought to be a result of their evolution on separate continents (Allende et

al., 1999; Nelsen et al., 1999). In addition, there is a great deal of genomic variation

within PRRS viruses isolated from within a continent and even within a given country or

state. PRRS viruses of the NA genotype isolated in Central China have been shown to

share as low as 88% nucleotide identity genome-wide and as low as 72% identity when

comparing individual genes (CAN et al., 2016). This heterogeneity is due to the error-

prone nature of the PRRSV RNA polymerase, recombination events and selective

pressure in the field (Murtaugh et al., 2010). For this reason, there continues to be genetic

and subsequently antigenic difference between PRRS virions, and this poses the greatest

challenge to vaccine developers today, and to the general control of the disease.

Both North American and European isolates have a nearly identical genome organization.

It consists of a linear positive sense, single stranded RNA genome 15 kb in length with

eight open reading frames (ORFs). ORF1a and b encode non-structural proteins involved

in replication, while ORF2-7 code for structural proteins. ORFs 2 to 4 encode the minor

structural glycoproteins GP2, GP3 and GP4, respectively while ORFs 5 to 7 encode the

major structural proteins glycoprotein 5 (GP5), membrane protein (M) and nucleocapsid

(N), respectively (Nelsen et al., 1999). Both GP5 and M dominate the virion’s surface

and form a heterodimeric structure that is important for virus assembly (Mardassi et al.,

1996; Verheije et al., 2002) (Figure 1).

3

Figure 1. Schematic showing the structure of the PRRS virion. M and GP5 dominate the

surface of the virion as a heterodimer while minor structural proteins GP2, GP3 and GP4

exist as a complex within the lipid envelope.

4

1.1.1 The two major envelope proteins of PRRSV

GP5 is the major glycosylated envelope protein of PRRSV and originates from

one of the most variable regions in the viral genome, the ORF5 gene. The ORF5 gene of

one PRRSV isolate can share as little as 84% nucleotide identity with that of another

isolate from the same genotype (Chen et al., 2006). ORF5 encodes the transmembrane

protein GP5 which is typically 200 amino acids long. The first 28-30 amino acids make

up the putative signal peptide, which is assumed to be cleaved to give rise to the mature

protein (Mardassi et al., 1996). Following this is the domain of GP5 that exists outside

the virus, the ectodomain, which is approximately 31 residues in length with two

potential N-glycosylation sites. After the ectodomain, there is a long hydrophobic region

of about 70 residues that spans the membrane three times, and finally a long hydrophilic

domain existing within the virus, the endodomain, of approximately 68 residues (Figure

2). PRRSV GP5 is a target of most of the neutralizing antibodies in an infected animal

(Dea et al., 2000), and this is not surprising because it lies exposed at the surface of the

virion and is highly abundant. These immunoreactive epitopes involved in virus

neutralization are ideal candidates for display on nanoparticles. With production of

neutralizing antibodies against PRRSV, uninfected pigs can be protected from viral

challenge and infection, while the viral load of infected pigs can be reduced. Using the

sera of PRRSV infected pigs and a series of overlapping peptides derived from the GP5

ectodomain of PRRSV VR-2332 strain, Plagemann and colleagues (2002) identified an

antibody binding site between amino acids 36 and 52. In agreement with this data, amino

acids 37 to 45 were identified as a conserved region that is reactive with pig sera

containing high levels of neutralizing antibodies, and is recognized by a known PRRSV-

neutralizing antibody (Ostrowski et al., 2002). Neutralizing antibodies have also been

shown to target residues 29-35 of the GP5 protein (Wissink et al., 2003) further

signifying the potential of the GP5 ectodomain as a target in my vaccine design.

The M protein is the other major structural protein found in the envelope of

PRRSV and it originates from the ORF6 gene (Mardassi et al., 1995). M is typically 174

amino acids long and has a similar membrane topology to GP5; it begins with an

ectodomain of about 18 residues, traverses the PRRSV membrane three times, and ends

5

with a C-terminal endodomain of about 72 amino acids (Figure 2). Unlike GP5, M is not

glycosylated and considered as the most conserved structural protein of PRRSV (Dea et

al., 2000). However, despite being abundant at the surface of the PRRS virion, there are

no identified neutralizing epitopes in the ectodomain of M. The large endodomain of M is

thought to be the most immunoreactive region of the protein, having B cell epitopes

within residues 150-174, however, none of these epitopes are involved in the

neutralization of the virus (de Lima et al., 2006). Jiang and colleagues showed that the

presence of M, although not inducing neutralizing antibodies against itself, can amplify

the immune response and production of neutralizing antibodies against GP5 when co-

expressed as a fusion protein (M-GP5) in mice (Jiang et al., 2006).

For PRRSV and two closely related arteriviruses, lactate-dehydrogenase-elevating

virus and equine arteritis virus, researchers demonstrated that GP5 and M homologs from

all three viruses respectively; coprecipitate when targeted by anti-GP5 and anti-M

monoclonal antibodies and comigrate on agarose gels in non-reducing conditions. Upon

exposure to reducing conditions, GP5 and M proteins are resolved during electrophoresis

indicating their interaction via a disulfide linkage (de Vries et al., 1995; Faaberg et al.,

1995; Mardassi et al., 1996). GP5-M heterodimers, formed by the disulfide bridge

between their ectodomains (Figure 2), are thought to be essential for virion assembly

(Snijder et al., 2003; Verheije et al., 2002).

6

Figure 2. Schematic diagram showing the predicted topology of the proteins M and GP5

of PRRSV VR-2332 within the virus envelope. Their dimerization via a disulfide bond is

shown.

7

1.2 Vaccines and vaccination approaches

Vaccines are biological agents that allow for the presentation of one or more recognizable

features of a disease-causing micro-organism (pathogen) to the immune system. Since the

late 18th century with the introduction of the first vaccine by Edward Jenner (Riedel,

2005), vaccines have played a pivotal role in the prevention of infectious diseases and

death. The ideal vaccine stimulates a strong immune response and leads to the

development of immunological memory cells that essentially remain in the recipients,

readily equipping them with the means to prevent disease if exposed to the recognizable

pathogen in the future.

Traditionally, vaccination approaches began with the development and administration of

chemically inactivated forms of a virus (termed killed vaccines), or by reducing the

virulence of a virus by passage in tissue culture or animal hosts (termed ‘live-attenuated’

vaccines). While these approaches have resulted in the development of many effective

vaccines, they have their drawbacks (Plotkin, 2014). Killed viruses do not tend to

stimulate a sufficiently strong immune response thus it is usually required that recipients

receive multiple doses and/or adjuvants to acquire desired immunity. Conversely, live-

attenuated viruses are typically very efficient immunological stimulants, but they

replicate in the host and can acquire a collection of mutations that has the potential to

cause reversion to a disease-causing virus. Another disadvantage associated with some

live-attenuated vaccines is that, despite their reduced virulence, these viruses continue to

produce proteins that modulate host cell responses to invading viruses, thus preventing an

adequate immune response (Renukaradhya et al., 2012).

With an obvious desire for improved vaccines, and an increased understanding of

immunobiology, in the early 20th century came the development of subunit vaccines

which contain purified or recombinantly-produced proteins from a pathogen. The

proteins from a pathogen that induce an immune response are commonly referred to as

antigens. While being safer than both killed and live-attenuated vaccines, subunit

vaccines lack the ability to effectively stimulate the immune system and possible reasons

for this include improper folding of the antigen, inadequate presentation to the immune

system and the general instability of the soluble antigens (Chua et al., 2011; Liu et al.,

8

2018). Like killed vaccines, more doses at higher concentrations of antigens and

adjuvants are necessary for protective immunity and thus subunit and killed vaccines are

considerably more expensive than live-attenuated vaccines (Liu et al., 2018; Noad and

Roy, 2003).

1.2.1 PRRSV vaccines

Currently there are two types of PRRSV vaccines commercially available, a live-

attenuated and a killed version. As mentioned previously, although successful as

vaccines, live-attenuated PRRSV vaccines carry risks associated with their use since

there is a potential for reversion to virulence (Botner et al., 1997; Plummer and

Manchester, 2011). As for the killed PRRSV vaccines, they are generally associated with

insufficient B cell activation, unsatisfactory viral clearance and in controlled experiments

they have been shown to provide no protection from disease upon viral challenge (Piras

et al., 2005; Renukaradhya et al., 2015). For this reason, killed PRRSV vaccines are no

longer used in the United States. Also, since both PRRSV genotypes are distributed

worldwide and their divergence can significantly affect vaccine efficacy, many vaccines

fail to provide cross-protective immunity; i.e. immunity against heterologous strains of

the PRRS virus (Lager et al., 1999; Martelli et al., 2009; Mengeling et al., 2003). With

these drawbacks in mind, there is a need for new and innovative vaccines for PRRS.

1.3 Virus-like particles

A major advance for subunit vaccines came in the late 20th century when Hepatitis B core

antigen (HBcAg), which forms the nucleocapsid of the hepatitis B virus, was expressed

in E. coli, purified and visualized by electron microscopy (Richmond and Cohen, 1982).

HBcAg is formed from 240 copies of a single capsid protein and Cohen and Richmond

(1982) were able to show that HBcAg self-assembled to form virus-like particles (VLPs)

that were indistinguishable from native Hepatitis B viral core and had comparable

antigenic properties. By definition, VLPs are composed of viral structural proteins that

9

assemble into nanostructures that ultimately resemble a virus. VLPs vary in their level of

complexity and HBcAg is an example of a simple VLP (Richmond and Cohen, 1982).

More complex VLPs involve the assembly of multiple copies of two or more different

structural proteins, for example, the VLP derived from the Salmonella typhimurium

bacteriophage P22 (VLP P22). VLP P22 is formed from 430 copies of a coat protein

along with 100-300 copies of a scaffolding protein (Yoshimura et al., 2016). Beyond

VLPs consisting of viral structural proteins alone, there are VLPs composed of viral

structural proteins within a lipid envelope (Figure 3). These envelope VLPs occur in

cases where viral proteins that are involved in budding of the native virus from infected

cells are recombinantly expressed. An example of such a protein is the Gag polyprotein

of the human immunodeficiency virus type 1 (Cervera et al., 2013).

Soon after the discovery of HBcAg, researchers took a similar approach to construct and

characterize various self-assembling viral proteins and found that several had very

attractive features for vaccine development (Christianson, 1992; McAleer et al., 1984;

Miyanohara et al., 1986; Thuenemann et al., 2013). They mimicked the structure of the

native infectious viruses; they lacked a viral genome (no capacity of self-replication) and

they elicited high antibody titers in animals (good immunostimulants). Immediately,

VLPs became very appealing candidates as vaccines for the virus from which they

originated especially since they also contained no viral proteins that could downregulate

the immune system. This meant that lower doses were expected to be necessary to mount

a desirable protective immune response. The potency of VLPs in stimulating and

sustaining the immune system to the point of long-lasting protection is largely due to the

dense repetitive structure of the exposed surface and their particulate nature (Chaplin,

2010).

The immune cells of the adaptive immune response, the B and T cells, are responsible for

continued protective immunity and they are both well stimulated upon exposure to VLPs.

The polymeric, repetitive surface of VLPs ensures multivalent presentation of epitopes

that induce B cells, and the particulate nature of VLPs encourages internalization by

antigen presenting cells that later activate T cells (Chaplin, 2010). In comparison,

vaccines based on free soluble antigens generally do not stimulate immune cells

10

adequately because they have weak, short-lasting interactions with immune cell

receptors. A VLP on the other hand, is an assembly of multiple copies of an antigen or

antigens that together provide multiple interactions at the surface of immune cells

cumulatively strengthening the interaction (López-Sagaseta et al., 2016). This long-

lasting interaction promotes effective intracellular signaling which leads to a strong

immune response and long-lasting immunity. To date, there are successful US FDA

approved VLP vaccines for human papillomavirus including Gardasil® (Merck and Co.

Inc.) and Cervarix® (GlaxoSmithKline) (Rodríguez-Limas et al., 2013) and VLP

vaccines in clinical trials for influenza virus (Roldão et al., 2010), norwalk virus (Roldão

et al., 2010) and chikungunya virus (Sun et al., 2010). Figure 3 summarizes the key

drawbacks of the different vaccines being used today.

While VLPs have a strong potential for presenting antigens to the immune system in a

structured, repetitive and safe format, producing VLPs in general still poses an issue.

Unlike HBcAg, where self-assembly is induced simply by increasing salt concentration

(Bundy et al., 2008), there are many cases where VLPs require additional biomolecules

such as RNA or even additional proteins to induce their assembly. One such example is

the VLP made from the coat protein of the bacteriophage MS2. These bacteriophage

VLPs consist of 180 copies of a single coat protein that, in native conditions, hold the

RNA genome. MS2 bacteriophage VLPs have been successfully used as platforms for a

multitude of molecules including antibody fragments, glycoproteins and nucleic acids,

however, their assembly relies on a specific stem-loop structure found in the RNA

genome (Patel and Swartz, 2011). Production of MS2 VLPs is typically a strenuous

procedure involving culturing the virus in E. coli, purification of the virus itself,

disassembly of the virus using denaturant, careful precipitation and removal of genomic

RNA, removal of the denaturant by dialysis, then reassembly with the addition of the

stem-loop RNA (Ashley et al., 2011). Another issue that arises with the use of VLPs is

that some VLPs have restrictions on where they can be produced; not all VLPs can be

produced in E. coli and yeast, the most common protein expression systems. Cow pea

mosaic virus VLPs for example must be produced in plant or insect cells where its initial

coat protein polypeptide can be properly processed by a proteinase into its component L

and S coat proteins (Saunders et al., 2009).

11

The difficulty in assembling most VLPs, compounded with the fact that there are

limitations on what molecules can be accommodated on any one specific VLP,

encouraged the study of nanostructures based on non-viral proteins. Exploring non-viral

protein nanoparticles as platforms opened up the possibility of finding more stable and

versatile platforms that can carry antigenic peptides that could not be produced as VLPs

previously (Frietze et al., 2016). Today, researchers use many different protein

nanoparticles as platforms or cages to display chosen antigens in a polymeric and

repetitive fashion similar to VLPs, or entrap antigens for delivery (Dalmau et al., 2008;

Domingo et al., 2001; Flenniken et al., 2003).

12

Figure 3. Summary of the different types of vaccines highlighting key drawbacks of

each.

13

14

Figure 4. Surface representation of the candidate antigen carriers. A. Brucella lumazine

synthase (PDB code: 1t13) highlighting from left to right its monomers within the

pentamer, the dimer of pentamers and the N-termini (green) of all 10 monomers. B.

Aquifex aeolicus lumazine synthase (PDB code: 1nqx_1) highlighting from left to right

its five monomers within the spherical structure, the C-termini (red) of the five monomers

and the C-termini of all 60 monomers. C. Helicobacter pylori ferritin (PDB code: 3bvf)

highlighting from left to right, two monomers within the spherical structure, the N-

termini (green) of the two monomers and the N-termini of all 24 monomers. D.

Methanococcus jannaschii small heat shock protein (PDB code: 4i88) highlighting from

left to right four monomers within the spherical structure, the C-termini (red) of the four

monomers and the C-termini of all 24 monomers. Scale bar: 5 nm. All imaged produced

in PyMOL.

15

1.4 Platforms of interest

My goal is to present antigens from porcine reproductive and respiratory syndrome virus

(PRRSV; described below) in a multimeric virus-like fashion using self-assembling

protein nanoparticles. The nanoparticles to be tested include lumazine synthase from

Brucella, lumazine synthase from Aquifex aeolicus, ferritin from Helicobacter pylori and

the small heat shock protein from Methanococcus jannaschii (Figure 4).

1.4.1 Brucella lumazine synthase

Lumazine synthase (LS) is a polymeric bacterial enzyme involved in riboflavin synthesis

and exists in various quaternary forms depending on the organism from which it

originates. LS isolated from Brucella spp. (BLS) can be found in its more commonly

known pentameric form (Braden et al., 2000) or as a stable dimer of pentamers joined

head to head (Zylberman et al., 2004) (Figure 4A). BLS is well studied and has been

shown to be an effective immunostimulant and good candidate for use in vaccination

against brucellosis since it can generate both a strong humoral and cell-mediated immune

response even in the absence of adjuvants which generally accompany subunit vaccines

to establish efficacy (Rossi et al., 2015; Velikovsky et al., 2003, 2002). Structural

analysis and previous work on BLS have shown that it is highly resistant to chemical and

thermal denaturation and it contains a disordered N-terminus of around 10 amino acids

that can be replaced with foreign peptides with no apparent effect on the folding

capabilities or stability of the decamer (Bellido et al., 2009; Cassataro et al., 2007;

Zylberman et al., 2004) (Figure 4A). As such, it has become a prominent protein carrier

of foreign peptides in vaccine development and other biomedical applications (Laplagne

et al., 2004; Rosas et al., 2006).

1.4.2 Aquifex aeolicus lumazine synthase

LS isolated from the hyperthermophile Aquifex aeolicus (AaLS) exists as a hollow

icosahedral structure with a diameter of about 15.4 nm made up from 12 LS pentamers

(Zhang et al., 2001) (Figure 4B). AaLS has predominantly been used as a cargo system

whereby negatively or positively charged amino acid residues are introduced to the

interior of the cage to accommodate oppositely charged cargo via electrostatic attraction.

16

A few examples of such cargo include proteins such as HIV protease (Worsdorfer et al.,

2011) and “supercharged” GFP (Worsdorfer et al., 2012), nucleic acids (Lilavivat et al.,

2012) and anticancer drugs (Wang et al., 2018). To target drug delivery to certain cells,

Min and colleagues successfully added cell targeting peptides to both the C-terminus of

AaLS, and within a loop structure of the AaLS monomer that is surface exposed in the

final icosahedral structure (Min et al., 2014). By doing this, they showed that AaLS can

accommodate foreign peptides in these exterior positions without altering its overall

architecture.

1.4.3 Ferritin

Ferritins are a family of globular iron storage proteins that maintain iron homeostasis in

organisms from all kingdoms. Free iron is toxic to cells since it participates in the

production of harmful reactive oxygen species. Ferritin prevents this by sequestering and

converting the harmful ferrous (Fe2+) form of iron to its ferric (Fe3+) form which is safely

compartmentalized as iron (III) oxide (Jameson et al., 2002). Structurally, ferritin is a 24-

subunit hollow nanoparticle with a diameter of about 12 nm; dimers form initially and

later self-assemble into a dodecameric cage (Cho et al., 2009) (Figure 4C). This cage is

thermostable up to 85°C, structurally sound in a wide pH range (3.4 to 10) and resistant

to denaturation in relatively high levels of urea and guanidinium at neutral pH (Kim et

al., 2011; Linder et al., 1989; Otsuka et al., 1980). Over the years the surface of ferritin

has been extensively used as a platform for several molecules including fluorescent dyes,

targeting peptides and antibodies (Truffi et al., 2016). This functionalization of ferritin is

performed either by chemical conjugation or genetic fusion. Ferritin has been shown to

accommodate genetic fusions of foreign peptides at both its termini and in the middle of a

flexible loop exposed on its surface (Kanekiyo et al., 2013; H. J. Kang et al., 2012; Y. J.

Kang et al., 2012). An N-terminal genetic fusion protein of ferritin has previously been

produced whereby the influenza virus hemagglutinin was fused to the surface of the

Helicobacter pylori ferritin resulting in a more potent influenza vaccine than one that was

commercially available (Kanekiyo et al., 2013).

17

1.4.4 Small heat shock protein

The small heat shock protein (sHSP) is a stress response protein that assists the folding of

proteins by stabilizing their folding intermediates. In the thermophilic archaeon,

Methanococcus jannaschii, sHSP exists as a 24-subunit hollow sphere with a diameter of

about 12.4 nm (Quinlan et al., 2013) (Figure 4D). This cage is stable up to 70°C, within a

pH range of 5 to 11 and amenable to both genetic and chemical modification (Bova et al.,

2002; Flenniken et al., 2006, 2005). Flenniken and colleagues (2006) engineered a cell-

targeting protein nanoparticle using M. jannaschii sHSP (MjHSP16.5) as a platform by

genetically fusing an integrin ligand to the C-terminus of the MjHSP16.5 monomer. The

presence of the ligand on each monomer did not affect assembly of the MjHSP16.5 cage;

the ligands were displayed on the cage’s exterior and were functional (Flenniken et al.,

2006). They also demonstrated that MjHSP16.5 is such a robust structure that it tolerates

the chemical conjugation of a functional, cell-targeting monoclonal antibody onto its

surface.

1.5 Rationale and goal

Previous work on these nanoparticles of interest demonstrates they are amenable to

genetic fusion enabling the creation of stable multivalent structures with a uniform

distribution of foreign peptides on their surfaces. The goal of my thesis is to develop a

novel PRRSV vaccine prototype via the genetic fusion of antigenic PRRSV peptides to

each of the aforementioned proteins.

1.6 Objectives

My objectives are to select a candidate peptide from the PRRS virion to display on the

protein nanoparticles of interest, design the fusions between it and the protein monomers,

express these recombinant fusion proteins in E. coli, identify ones that are well expressed

and can be purified, and finally characterize these fusions along with the unfused

nanoparticles. Figure 5 shows one example of the fusion proteins I have produced, a

surface representation of the extended form of the GP5 peptide displayed on the surface

of BLS.

18

Figure 5. Surface images of the Brucella lumazine synthase (BLS) fused to the extended

representation of the GP5 antigen. From left to right; side view of BLS, side view of

GP5-BLS fusion and top view of GP5-BLS fusion. Black, flexible linker (x4GGS);

green, GP5-antigen; red, purification tag (x6His-tag). Scale bar: 5 nm. All images

produced in PyMOL.

19

Chapter 2

2 Experimental Procedures

Genetic constructs were designed and sent to Gene Universal Inc. for synthesis within

expression plasmids to overexpress the self-assembling proteins being tested, either fused

or unfused to an antigen from PRRSV. These expression plasmids were transformed into

Escherichia coli and the various proteins were overexpressed, purified and characterized.

2.1 Construction of chimeras

Amino acids 1 to 60 from the GP5 protein of 110 North American PRRSV strains were

aligned to reveal a well conserved region in the ectodomain of GP5. The selected GP5

amino acid sequence was NASNDSSSHLQLIYNLTLCELNGTD, corresponding to the

well conserved region (amino acids 30-54) of the GP5 protein (accession number:

AAO13196.1) from PRRSV strain VR-2332. The selected M amino acid sequence was

MGSSLDDFCHDSTAPQKV, corresponding to the entire first ectodomain (amino acids

1-18) of the M protein (accession number: AAO13196.1) from PRRSV strain VR-2332.

The flexible linker chosen was a tetrapeptide repeat of GGS and amino acids 8- 158 of

BLS (PDB code: 1T13) was used as the BLS monomer. Amino acid sequences were

converted to DNA sequences and optimized by Gene Universal Inc. for protein

expression in E. coli.

Fused chimeras were synthesized as either N-terminal fusions (GP5-linker-BLS, GP5-

linker-Ferritin, M-linker-GP5-linker-BLS) or C-terminal fusions (HSP-linker-GP5,

AaLS-linker-GP5); refer to Figure 6. N-terminally fused constructs were cloned into the

expression plasmid pET-28a(+) while C-terminally fused constructs were cloned into

pET-24a(+) by Gene Universal Inc. Six histidine residues were fused to the N-terminal or

C-terminal end of each of the proteins during cloning into pET-28a(+) or pET-24a(+),

respectively. BLS was kindly donated by Fernando Bravo-Almonacid (Alfano et al.,

2015), amplified using oligonucleotide primers CPG010F (5’-

TACTTCCAATCCAATGCAAAGACATCCTT -TAAAATCGC-3’) and CPG010R

(5’F-TTATCCACTT- CCAATGTTATTAGACAAG -CGCGATGCGGCTGCG-3’R),

20

isolated and cloned into pET His6 TEV LIC cloning vector (1G); a gift from Scott Gradia

(Addgene plasmid # 29655). Site-directed mutagenesis using primers OHH017F (5’-

GCGCAGCCGCATCG-CGGCCCTTGTCT -AATAACATTGG-3’) and OHH017R (5’-

CCAATGTTATTAGACAAGGGCCGCGA -TGCGGCTGCGC-3’) was performed to

replace a missing alanine at amino acid position 156 that is native to BLS.

21

Figure 6. Schematic of the designed GP5-fusions. From left to right on the N-terminal

fusion, the 6x-histidine tag, GP5 amino acids 30-54, 4x GGS linker, monomer of

nanoparticle. BLS and ferritin were designed as N-terminal fusions while HSP and AaLS

were designed as C-terminal fusions.

22

2.2 Expression and purification

2.2.1 Expression and lysis

All expression plasmids were transformed into BL21 (DE3) E. coli cells (NEB)

and one colony was grown overnight at 37°C in 5 ml of Luria-Bertani broth containing

50 µg/mL kanamycin (LBkan). The overnight culture was used to inoculate a 500 ml

culture grown the following day. After an OD of 0.8-1.0 was reached, protein expression

was induced with 0.5 mM Isopropyl β-D-1-thiogalactopyranoside (IPTG) for 3 hours at

37°C, with shaking at 250 rpm. The induced cells were harvested by centrifugation using

the Sorvall RC5B Plus at 5,465 x g for 10 minutes at a temperature of 4°C.

Cells expressing wild-type recombinant proteins were resuspended in buffer A

(50 mM Tris-HCl (pH 8.5), 500 mM NaCl) while those expressing chimeras were

resuspended in buffer B (50 mM Tris-HCl (pH 8.5), 500 mM NaCl, 8 M urea, 14.3 mM

2-mercaptoethanol). In the following steps, 14.3 mM 2-mercaptoethanol was present in

buffers used to process all engineered recombinants if not stated otherwise. Lysis was

performed by sonication (QSonica) at 30W for 5 minutes, using 30/30 seconds on/off

rounds. After centrifugation at 20,000 x g for 30 minutes at 4°C, the cleared cell lysate

was collected, and imidazole was added to a final concentration of 5 mM.

2.2.2 Immobilized metal affinity chromatography

Immobilized metal affinity chromatography (IMAC) was performed to enrich for

His6-tagged proteins since histidine residues have a high affinity for the Ni2+-charged

resin in the chromatography column. IMAC was performed by passing the cleared cell

lysate through 5 mL Ni Sepharose 6 Fast Flow (GE Healthcare) resin in the column then

washing with buffer A or buffer B containing first 5 mM imidazole, then 25 mM

imidazole. The proteins were eluted in buffer A or B containing 100mM NaCl and 500

mM imidazole. Proteins were analyzed by SDS-PAGE using a 4-20% gradient

polyacrylamide gel (Bio-Rad), a Tris-Tricine running buffer (Bio-Rad) and visualized by

staining the gel with Coomassie Blue. Eluates containing the proteins of interest were

23

pooled and dialyzed thoroughly against 50 mM Tris-HCl (pH 8.5), 100 mM NaCl.

However, due to the amount of precipitate formed in the dialysis buffer used for the other

chimeras (50 mM Tris-HCl (pH 8.5), 100 mM NaCl, 14.3 mM 2-mercaptoethanol ), the

eluate of the chimera GP5-HSP was dialyzed repeatedly in 50 mM Tris-HCl (pH 8.5),

100 mM NaCl, 2 mM dithiothreitol (DTT); 2 mM DTT was kept in the buffers in future

steps.

2.2.3 Anion-exchange chromatography

The dialyzed samples were further enriched by anion-exchange chromatography

using an NGCTM Quest plus Fast-Performance Liquid Chromatography (FPLC) system

(Bio-Rad) and either an Enrich mono Q 5 x 50 column (Bio-Rad) or HiTrap Q HP 5 x 1

ml column (GE Healthcare). Columns were equilibrated with 50 mM Tris-HCl (pH 8.5),

100 mM NaCl while the elution buffer was 50 mM Tris-HCl (pH 8.5), 1 M NaCl.

Typically, runs were carried out at 1 ml/min with a gradient of 0-100% elution buffer

over 25 ml and 0.45 ml fractions were collected. Eluted fractions were analyzed by SDS-

PAGE and fractions containing the desired protein were pooled and further purified by

size-exclusion chromatography.

2.2.4 Size-exclusion chromatography

Size-exclusion chromatography (SEC) was performed to further purify the

proteins and acquire an average molecular weight for the self-assembled protein

structures. An ENrich SEC 650 10 x 300 mm column (Bio-Rad) equilibrated in 50 mM

Tris-HCl (pH 8.5), 100 mM NaCl was used.

24

2.3 Characterization

2.3.1 Transmission electron microscopy

Pure protein (5 ul) following size-exclusion chromatography was dropped onto

400 mesh copper grids coated with formvar carbon film (Electron Microscopy Sciences),

washed with 50 mM Tris-HCl (pH 8.5), 100 mM NaCl then negatively stained for one

minute with 2% uranyl acetate. The structure of the purified nanoparticles was imaged

using the Jeol JEM-1200EXII transmission electron microscope with an accelerating

voltage of 60 kV.

2.3.2 Circular dichroism spectroscopy

Circular dichroism (CD) spectroscopy was carried out at the Biomolecular

Interactions and Conformations Facility at Western University. Pure BLS at a

concentration of 0.79 mg/ml and pure M-GP5-BLS concentrated using Vivaspin 500 (GE

Healthcare) to a concentration of 0.77 mg/ml were used for CD analysis. CD spectra were

obtained on a Jasco J-810 spectropolarimeter using a 0.1 mm path length quartz cell.

Scans were recorded from 260 to 190 nm with a step size of 0.5 nm, a scanning speed of

100 nm/min, a response time of 0.5 s, a data pitch of 0.5 nm, and a band width of 1 nm at

20 °C. To reduce background noise, fifteen spectra were recorded for each sample and

the average spectrum obtained as raw data. Buffer scans were subtracted from the raw

data and corrected values were converted to molar ellipticity using the concentration

values determined from their absorbance at 280 nm measured on a NanoDrop One

spectrometer (Thermo Scientific). The melting temperature (Tm) of individual constructs

was monitored at 222 nm using a temperature gradient from 50 °C to 105 °C using a step

size of 1 °C at a temperature change rate of 1 °C/min. To determine the Tm, the data were

fitted by non-linear regression to the equation:

𝑦𝑜𝑏𝑠 =

(𝑦𝑛 + 𝑚𝑛𝑇) + (𝑦𝑑 + 𝑚𝑑𝑇) (𝑒𝛥𝐻𝑚

𝑅×(

1𝑇𝑚

−1𝑇

))

1 + 𝑒𝛥𝐻𝑚

𝑅×(

1𝑇𝑚

−1𝑇

)

25

(1)

In Equation 1, 𝑦𝑜𝑏𝑠 is the absorbance observed during the experiment, 𝑦𝑛 and 𝑦𝑑 are the

y-intercepts of the native and denatured baselines respectively, 𝑚𝑛 and 𝑚𝑑 are the slopes

of the native and denatured baselines respectively, T is the temperature in degrees Kelvin,

Tm is the melting temperature, R is the gas constant and 𝛥𝐻𝑚 is the enthalpy of

unfolding.

2.3.3 Sedimentation velocity

Sedimentation velocity studies were performed to provide information about the

molecular weight and shape of the purified nanoparticles. These studies were carried out

at the Biomolecular Interactions and Conformations Facility at Western University, using

a Beckman Optima XL-A Analytical Ultracentrifuge. Samples were loaded into a double-

sector cells with Epon charcoal centerpieces and centrifugation was carried out at 20 °C

using an An60Ti rotor. Absorbance was monitored at 280 nm and scans were taken every

10 minutes for a total of 45 scans, in 0.002 cm radial steps. BLS was sedimented at a

speed of 30,000 rpm and M-GP5-BLS at 25,000 rpm. Three samples of BLS and M-GP5-

BLS at various concentrations were analyzed by sedimentation velocity in buffer

containing 50 mM Tris-HCl and 250 mM NaCl. Prior to sedimentation, for both BLS and

M-GP5-BLS, concentrations were adjusted to 0.3, 0.6 and 0.9 mg/ml assuming an OD280

of 1 is equal to 1 mg/ml; measured on the NanoDrop One (Thermo Scientific). Partial

specific volumes (vbar) of BLS (vbar = 0.740) and M-GP5-BLS (vbar = 0.727) were

calculated from their amino acid compositions using the program SEDNTERP. Buffer

density (1.0099 g/mL) and viscosity (0.0104 Poise) were also calculated with

SEDNTERP. Sedimentation data were analyzed using the c(s) distribution model in

Sedfit and frictional ratios for BLS (f/f0= 1.3) and M-GP5-BLS (f/f0=1.5) were estimated

by non-linear regression in order to get the best-fit c(s) distribution. Data were

normalized in GUSSI and exported to GraphPad Prism for figure making.

26

Chapter 3

3 Results

3.1 The GP5-antigen chosen is conserved and immunogenic

When selecting a PRRSV peptide sequence to genetically fuse to the nanoparticles of

interest, there are several characteristics I sought to increase the likelihood of obtaining

an effective vaccine. Ideally, the peptide should be immunogenic, abundantly displayed

on the surface of the virus, and conserved across strains. An immunogenic peptide will

effectively induce the immune response and increase the likelihood of long-lasting

protection against PRRSV. A peptide that is abundantly displayed on the surface of the

virion will provide multiple binding points for antibodies and increase the chances of

PRRSV neutralization upon infection. Finally, a peptide sequence that is conserved,

common to many strains of a virus, is likely to induce the production of antibodies that

will target this sequence in all the different strains of the virus in which this target

sequence lies. Therefore, a conserved, immunogenic and surface exposed peptide is ideal

in my vaccine design as it will have the greatest potential to provide protection against

multiple strains, and potentially genotypes (North American and European) of the virus.

I selected a peptide from the ectodomain of the GP5 protein because it dominates the

surface of the virion, and within its ectodomain lies an immunogenic and conserved

region (Mardassi et al., 1996; Ostrowski et al., 2002; Plagemann et al., 2002). Figure 7

illustrates the conservation present within the PRRSV GP5 ectodomain from both

genotypes. One-hundred and ten isolates from each genotype (NA and EU) were aligned

and results showed that within each genotype, the ectodomain of GP5 is well conserved.

Amino acids 40-56 of the NA genotype and 42-58 of the EU genotype have the

consensus sequence Q(L/Y)IYNLT(I/L)CELNGTDWL, whereby only two amino acids

in their sequences usually differ. This strongly indicates conservation of the GP5

ectodomain between genotypes. Figure 7 also shows that the putative transmembrane

signal peptide, the first 28-30 amino acids of GP5 that are cleaved from the mature GP5

protein (Mardassi et al., 1996), contains many hydrophobic amino acids.

27

I chose amino acids 30-54 of the GP5 membrane protein from the NA genotype which

excludes the hydrophobic signal peptide and includes the well conserved region present

within the NA genotype and the conserved region between both genotypes.

28

Figure 7. WebLogo (Crooks et al., 2004; Schneider et al., 1990) representation of the

multiple sequence alignment of the GP5 ectodomain of the PRRS virus from the North

American genotype and the European genotype. 110 isolates were aligned for both

genotypes and the height of the letters indicates the sequence conservation at that

position. Hydrophobic amino acids are colored black, charged blue and neutral green.

The chosen GP5-antigen is boxed in red.

29

3.2 Nanoparticles fused with the GP5-antigen are produced but are insoluble

Initial screening of the non-engineered nanoparticles showed robust over-expression

following induction with IPTG and most of these nanoparticles were soluble post-lysis

(Figure 8; BLS, sHSP 16.5, Ferritin). Lumazine synthase from Aquifex aeolicus (gel not

shown) overexpressed well, however, it was almost completely insoluble. Conversely, the

engineered nanoparticles, bearing GP5 genetically fused at specific termini express,

however, all were insoluble post lysis as indicated by the presence of the protein

exclusively in the pelleted fraction (Figure 8; GP5-BLS, sHSP 16.5-GP5, GP5-Ferritin).

To resolubilize these engineered constructs, I added 8M urea to the lysis buffer. With this

addition, a significant portion of the nanoparticle chimeras were resolubilized (Figure 9).

30

Figure 8. Expression and solubility of the wild-type or GP5 engineered protein

nanoparticles of interest. Coomassie blue-stained SDS-PAGE gel analysis of each

nanoparticle overexpressed in E. coli BL-21 (DE3) cells with 0.5 mM IPTG for 3 hours

at 37°C and lysed in TBS (50 mM Tris-HCL (pH 8.5), 500 mM NaCl). Pr, pre-induction;

Po, post-induction; Sup, supernatant; Pell, pellet; BLS, Brucella lumazine synthase; sHSP

16.5, small heat shock protein. Red arrows highlight the monomer that ran on the gel at

its expected size (BLS, 18.8 kDa; GP5-BLS, 22.4 kDa; sHSP 16.5, 17.5 kDa; sHSP 16.5-

GP5, 21 kDa; Ferritin, 21.6 kDa; GP5-Ferritin, 25.1 kDa).

31

Figure 9. Expression and solubilization of engineered nanoparticles with urea.

Coomassie blue-stained SDS-PAGE gel analysis of each chimera overexpressed in E. coli

BL-21 (DE3) cells with 0.5 mM IPTG for 3 hours at 37°C and lysed in TBS-urea (50

mM Tris-HCL (pH 8.5), 500 mM NaCl, 8 M urea, 14.3 mM 2-mercaptoethanol). Pr, pre-

induction; Po, post-induction; Sup, supernatant; Pell, pellet; BLS, Brucella lumazine

synthase; sHSP 16.5, small heat shock protein. Red arrows highlight the monomer that

ran on the gel at its expected size (GP5-BLS, 22.4 kDa; sHSP 16.5-GP5, 21 kDa; GP5-

Ferritin, 25.1 kDa).

32

3.3 Nanoparticles are enriched and partially purified by immobilized metal-affinity chromatography

All soluble recombinant proteins were initially purified by immobilized metal affinity

chromatography (IMAC). IMAC allows for the enrichment of histidine-tagged

recombinant proteins since the polyhistidine tag (usually a hexahistidine tag) has a high

affinity for the resin when charged with nickel ions. The interaction between the

polyhistidine tag and these immobilized metal ions is independent of the protein’s fold

and the fold of the tag itself. Since there is no specific conformation needed, this allows

for purification of the engineered constructs to occur under denaturing conditions such as

8 M urea. The nickel-charged nitrilotriacetic acid (Ni-NTA) resin binds His-tagged

proteins with high affinity but interacts poorly with other non-tagged proteins.

The IMAC step shows that wild-type nanoparticles can be enriched, with BLS and sHSP

16.5 showing the greatest enrichment, followed by ferritin where a good portion was

found in the flow-through and thus did not bind (Figure 10A). As for the chimeras, there

is a clear enrichment of sHSP-16.5-GP5 and GP5-Ferritin even though some of each is

lost during the process to the flow-through and the washes (Figure 10B); perhaps as a

consequence of the high concentration of urea or the protein was not properly

resolubilized to expose the His-tag. GP5-BLS in general did not express well (Figure 8)

and was difficult to enrich via IMAC (Figure 10B).

To refold the engineered constructs following IMAC, each one was dialyzed exhaustively

in buffer containing no urea. Figure 11 shows that a significant proportion of each

construct was resolubilized following dialysis to remove urea. However, a large fraction

of each construct could not be resolubilized. In an effort to increase the yield of my

potential vaccine candidates, constructs were redesigned taking into account evidence

that shows that the ectodomain of the PRRSV GP5 protein interacts with that of the M

protein ectodomain via a disulfide linkage (de Vries et al., 1995; Faaberg et al., 1995;

Mardassi et al., 1996).

33

Figure 10. Immobilized metal affinity chromatography. Coomassie blue-stained SDS-

PAGE gel analysis of fractions collected during the purification of the soluble fraction of

each cell lysate. A. Purifications in the absence of urea. B. Purifications in the presence

of 8M urea. S, sample loaded; FT, flow-through; 5, 25 and 500 refer to the mM

concentration of imidazole present in the buffers used for washing and eluting the His-

tagged protein from the Ni2+ resin. Each numbered lane corresponds to a 5 ml wash or

elution fraction. BLS, Brucella lumazine synthase; sHSP 16.5, small heat shock protein.

Red arrows direct you to the monomer which ran on the gel at its expected size (BLS,

18.8 kDa; GP5-BLS, 22.4 kDa; sHSP 16.5, 17.5 kDa; sHSP 16.5-GP5, 21 kDa; Ferritin,

21.6 kDa; GP5-Ferritin, 25.1 kDa).

34

Figure 11. Refolding of engineered nanoparticles. Coomassie blue-stained SDS-PAGE

gel analysis of the supernatant and pelleted fractions post-centrifugation of each dialyzed

sample. Elutions from the nickel purifications of each protein were pooled and dialyzed

against 50 mM Tris-HCl, 100 mM NaCl with 14.3 mM 2-mercaptoethanol (GP5-BLS

and GP5-Ferritin) or 2 mM dithiothreitol (HSP-16.5-GP5). BLS, Brucella lumazine

synthase; sHSP 16.5, small heat shock protein; sup, supernatant; pell, pellet.

35

3.4 The new M-GP5-BLS construct is an improvement on GP5-BLS

Previous research indicates that the structural proteins M and GP5 form a complex in the

outer-envelope of the virus, and this complex is assisted via a disulfide bond between

their solvent-exposed ectodomains (Mardassi et al., 1996) (Figure 2). It is distinctly

possible that this interaction creates unique antigenic epitopes not present in their

individual linear sequences. Unfortunately, a disulfide bond interaction in E. coli is

unlikely to be replicated since the cytoplasm of the bacterium is generally maintained as a

reducing environment where disulfide bonds are uncommon. However, if this interaction

is favorable and beneficial to the overall structure and solubility of the ectodomains of

these proteins, the presence of the M ectodomain paired with my chosen GP5 antigen

may improve refolding of the engineered particles in a non-reducing environment.

Therefore, in an attempt to maximize the amount of properly folded engineered

nanoparticles acquired that display antigenic epitopes in a more natural manner, I

designed two constructs using BLS as the carrier; M-GP5-BLS and GP5-M-BLS. BLS

was chosen as the carrier for the new antigenic epitopes because compared to GP5-

Ferritin and HSP-GP5, GP5-BLS displayed greater homogeneity and stability during

size-exclusion chromatography (SEC). SEC results of GP5-Ferritin and HSP-GP5

showed multiple elution peaks containing those proteins indicating greater dissociation of

their monomers compared to GP5-BLS, which generally eluted in one homogeneous peak

(data not shown).

In the case of M-GP5-BLS, starting from the N-terminus of the fusion protein, the first

ectodomain of the M protein, linked to the chosen antigenic GP5 sequence by a flexible

linker, and finally, GP5 is linked to the N-terminus of BLS via the same flexible linker

(Figure 12). With the GP5-M-BLS construct, the amino acid sequence of GP5 and M is

switched to test if the orientation of their amino acid sequences may affect their

interaction via the disulfide bond. The first ectodomain of M was chosen from the M

protein of the same PRRSV strain we selected the GP5- antigen (PRRSV strain VR-

2332).

36

With these new constructs synthesized, transformed into E. coli BL-21 cells and

expression induced, M-GP5-BLS expressed very well while GP5-M-BLS did not (Figure

13A). Upon lysis of these cells in non-denaturing conditions, it was evident that similarly

to all the other engineered nanoparticles, they were insoluble (Figure 13B) and thus

misfolded to form inclusion bodies within the E. coli cells. Therefore, to retrieve these

proteins, urea was included in the lysis buffer prior to IMAC. Analysis of the eluted

fractions from the nickel column shows that M-GP5-BLS eluted with high purity while

GP5-M-BLS was noticeably less pure in comparison. While there was one prominent

protein in the elution of M-GP5-BLS that ran on the SDS-PAGE gel at the expected MW

(25,2 kDa), there were three prominent protein species in that of GP5-M-BLS (Figure

13C). The highest prominent band on the GP5-M-BLS SDS-PAGE gel ran at the

expected size of GP5-M-BLS (25.3 kDa) while the second highest around 23 kDa and the

smallest, around 18 kDa (Figure 13C). These lower molecular weight proteins are

suspected to be degradation products of GP5-M-BLS and therefore M-GP5-BLS appears

to be the more favorable and stable orientation in the context of BLS as the carrier.

M-GP5-BLS was then dialyzed exhaustively against buffer without urea or reducing

agent (50 mM Tris-HCl (pH 8.5), 100 mM NaCl). Analysis of the supernatant and pellet

after removal of urea shows that a large fraction of M-GP5-BLS refolded and regained

solubility (Figure 13D). When dialyzed in a reducing environment a larger fraction of M-

GP5-BLS recovered its solubility upon removal of urea (Figure 13D). Therefore, there

was no clear indication that a disulfide bridge between the GP5 antigen and first

ectodomain of the M protein aided in the refolding of this new construct. However, there

was a clear improvement in expression, purification and dialysis of M-GP5-BLS

compared to the GP5-BLS construct. There was greater expression of M-GP5-BLS, its

purity following IMAC was markedly improved and upon removal of urea, a majority of

M-GP5-BLS remained soluble unlike GP5-BLS. M-GP5-BLS was chosen as my vaccine

candidate for future purification and characterization.

37

Figure 12. Schematic of the designed M-GP5-BLS construct, from left to right the 6x-

hisitidine tag, M protein amino acids 1-18, 4x GGS linker, GP5 protein amino acids 30-

54, 4x GGS linker, BLS amino acids 8-158.

38

39

Figure 13. Coomassie blue-stained SDS-PAGE gel analysis of the expression, solubility,

nickel purification and refolding of M-GP5-BLS and GP5-M-BLS. A. Each chimera was

overexpressed in E. coli BL-21 (DE3) cells with 0.5 mM IPTG for 3 hours at 37°C; Pr,

pre-induction; Po, post-induction. B. Supernatant versus pelleted fraction of cell lysates

lysed in in TBS (50 mM Tris-HCL (pH 8.5), 500 mM NaCl); Sup, supernatant; Pell,

pellet. C. IMAC of the soluble fraction of each cell lysate obtained under denaturing

conditions; S, sample loaded; FT, flow-through; 5, 25 and 500 refer to the mM

concentration of imidazole present in the buffers used for washing and eluting the His-

tagged protein from the Ni2+ resin. D. Supernatant and pelleted fractions of each sample

dialyzed against either 50 mM Tris-HCL (pH 8.5), 100 mM NaCl, 14.3 mM 2-

mercaptoethanol (reducing) or 50 mM Tris-HCL (pH 8.5), 100 mM NaCl (non-reducing).

Red arrows points to the monomer which ran on the gel at its expected MW (M-GP5-

BLS, 25.2 kDa; GP5-M-BLS, 25.3 kDa).

40

3.5 Size-exclusion chromatography indicates multimeric assembly of BLS and M-GP5-BLS

Following IMAC and dialysis of BLS and M-GP5-BLS into low salt buffers, they were

further purified by anion-exchange chromatography using a salt gradient from 100 mM to

1 M NaCl. Fractions bearing wild-type and engineered constructs were pooled and

concentrated for analysis by size-exclusion chromatography. The expected decameric

molecular weights of BLS and M-GP5-BLS are 188 kDa and 252 kDa respectively.

The elution profile of BLS shows a single, monodisperse peak with an elution volume of

13 ml corresponding to an apparent MW of approximately 172 kDa. BLS was the only

protein in the eluted fraction as indicated by SDS-PAGE (Figure 14). The elution profile

of M-GP5-BLS contains a prominent peak of pure M-GP5-BLS at 11.6 ml,

corresponding to an apparent MW of 505 kDa. Additional peaks and valleys were also

observed (Figure 14), however, SDS-PAGE analysis revealed no additional proteins were

present. The over-estimated MW of the prominent protein species in the M-GP5-BLS

sample may be due to the shape of the M-GP5-BLS decamer (See section 4.2).

41

Figure 14. Size-exclusion chromatography. Elution profiles of BLS (black line) and M-

GP5-BLS (blue line) monitored at a wavelength of 280 nm. Insets of SDS-PAGE gels

indicate purity of each peak with a monomer migrating at the appropriate size; 25.2 kDa

for M-GP5-BLS and 18. 8 kDa for BLS. S represents each sample prior to size-exclusion

chromatography. Elution volumes of the protein standards used to calibrate the size-

exclusion column are indicated above the chromatogram (670 kDa, thyroglobulin; 158

kDa, bovine γ globulin).

42

3.6 Sedimentation velocity analysis indicates that M-GP5-BLS is elongated while BLS is globular

Sedimentation velocity (SV) is an analytical ultracentrifugation method whereby

molecules in solution are subjected to a high centrifugal force. The rate at which they

move in response to this force is measured and provides information about the molecular

weight (MW) and shape of those molecules. Therefore, SV experiments were performed

to characterize and compare the shape and MW of BLS and M-GP5-BLS. The

sedimentation coefficient c(s) distribution of BLS standardized to conditions

corresponding to pure water at 20C (s20,w) shows a singular, sharp peak at an s20,w value

of 8.1 S (Figure 15A). This monodispersed distribution, with no other apparent peaks,

indicates that BLS forms a single structure that is stable and well-behaved in 50 mM

Tris-HCl (pH 8.5), 250 mM NaCl at 20°C, with no formation of aggregates. With this

type of c(s) distribution, a proper molecular mass distribution can be determined and with

that, the average molecular mass of BLS was calculated to be 178 kDa. This calculation

is a close approximation of the MW of decameric BLS whose expected MW is 188 kDa.

The frictional ratio of BLS was estimated to be 1.3 and this value informs on the globular

and symmetrical shape of the structure. In general, a perfectly spherical, compact and

smooth protein would have a minimum value of 1.0 and as the protein structure shifts

from globular to elongated, there is an increase in frictional ratio. Globular proteins

typically have frictional ratios ranging from 1.05 to 1.30 (Unzai, 2018).

The c(s) distribution of M-GP5-BLS in Figure 15B shows three peaks with the majority

of the engineered protein existing as a structure with the s20,w value of 8.8 S. This tells us

that the majority of M-GP5-BLS assembles into a structure larger than BLS, as indicated

by the larger s20,w value. The presence of multiple peaks indicates the presence of

aggregates in the cell. Although the c(s) distribution can be converted to molecular mass

distribution, with the existence of multiple peaks, calculation of the molecular mass of

each species at each peak is not expected to be accurate due to the assignment of one

frictional ratio to all species in one given run. However, the average molecular mass of

M-GP5-BLS calculated for the s20,w peak at 8.8 was 220 kDa. This calculation is

agreeable with the MW of decameric M-GP5-BLS whose expected MW is 252 kDa. As

43

for the shape of M-GP5-BLS, its frictional ratio of 1.5 indicates that it is not globular in

nature but rather an elongated protein structure.

44

Figure 15 Sedimentation coefficient distribution of BLS (A.) and M-GP5-BLS (B.).

Normalized sedimentation coefficient distribution, c(s), is plotted against the

sedimentation coefficient, s20,w. Sedimentation velocity experiments were conducted at an

initial protein concentration of 0.78 mg/ml (BLS) and 0.74 mg/ml (M-GP5-BLS) in 50

mM Tris-HCl (pH 8.5), 250 mM NaCl at 20°C. Data were collected at rotor speeds of

30,000 rpm and 25,000 rpm for BLS and M-GP5-BLS respectively. The calculated

values of the weight-average s20,w and frictional ratio of BLS are 8.1 and 1.3, respectively.

The calculated values of the weight-average s20,w and frictional ratio of M-GP5-BLS are

8.8 and 1.5.

45

3.7 Secondary structure and thermal stability are conserved in the M-GP5-BLS chimera

The far ultraviolet (UV)-circular dichroism spectra of BLS was compared with that of

both the BLS and chimeric M-GP5-BLS constructs which were extracted and processed

by IMAC in denaturing buffer (contained 8M urea) before being refolded by dialysis. All

steps following, anion-exchange chromatography and size-exclusion chromatography

were performed similarly to BLS not exposed to urea. The spectra of wild-type BLS

purified in the presence or absence of urea were virtually identical (Figure 16A.). Both

BLS spectra had two minima, one located at 220.5 nm and the other at 211 nm. As for the

chimera M-GP5-BLS, the overall shape of the spectra resembled that of BLS, with its

minima falling at 220 nm and 210 nm.

CDPro, a software package containing three different programs often used for secondary

structure assignment of proteins, facilitated the analysis of the CD spectra obtained for

each protein (Table 1). All analytical programs assigned a majority (>65%) of α- helical

content to BLS and BLS refolded with very little (<4.5%) β-sheet prediction, along with

about 15-20% unordered structure. These values are not what was expected since based

on crystal structures of BLS, it is 50-53% α- helical and 18-19% β-sheet (Kabsch et al.,

1983; Klinke et al., 2005; Zylberman et al., 2004). Since the α- helical content of BLS

was overestimated and β-sheet content underestimated, this indicates that the software

prediction used herein could not accurately discern the percentages of secondary structure

using the CD spectra obtained. However, assuming that α- helical content will be

overestimated and β-sheet content underestimated between different samples, changes in

secondary structure content between samples will still provide valid information.

The majority of M-GP5-BLS, like BLS, is also predicted to be α-helical however there is

a general decrease in the α-helical nature assigned to M-GP5-BLS compared to that of

BLS alone; with a 55% α-helical prediction being the highest assigned (Table 1

CDSSTR). M-GP5 consists of 87 amino acids including a His-tag, the first ectodomain of

PRRSV M protein, a 4x GGS linker, the chosen GP5 antigen followed by another 4x

GGS linker (Figure 12). Phyre2 secondary structure prediction software predicted the

secondary structure of M-GP5 to be 16% α- helical, 25% β-sheet and 60% disordered

46

(Kelley et al., 2015). This prediction agrees with the analytical programs that predicted an

increase in percentages of β-sheet, turn and disordered structure upon addition of M-GP5

to BLS. The changes are likely due to the addition of the antigen affecting the cumulative

absorbance of the BLS core structure.

To compare the thermal stability of BLS with that of the engineered counterpart, M-GP5-

BLS, their CD signals were monitored at 222 nm as a function of temperature. Figure 16

B and C show the thermal denaturation curves of BLS and M-GP5-BLS respectively. The

estimated melting temperature for BLS is 92.1 °C while that of M-GP5-BLS is 91.5 °C

which indicates that the addition of the antigen did not affect the thermal stability of the

carrier, BLS.

47

Figure 16. Circular dichroism (CD) spectroscopy of BLS and M-GP5-BLS. A. Far UV

CD spectra of BLS (solid line), BLS refolded (dashed line) and M-GP5-BLS (dotted

line). Results are representative of three experiments each with absorbances averaged

before conversion to mean residue ellipticity (MRE). B. CD signal of BLS measured at

222 nm as a function of increasing temperature; melting point estimated by non-linear

regression fit at 92.1 °C. C. CD signal of M-GP5-BLS measured at 222 nm as a function

of increasing temperature; melting point estimated by non-linear regression fit at 91.5 °C.

48

Table 1. Output summary from CDPro showing percentages of secondary structure

assigned to each protein by three different programs (CONTINLL, CDSSTR and

SELCON3) using the CDPro protein reference set SMP56.

CONTINLL Secondary structure assignment

α- helix (%) β-sheet (%) Turn (%) Unordered (%)

BLS 66.1 4.6 9.2 20.1

BLS refolded 68.7 3.4 8.5 19.4

M-GP5-BLS 48 10.2 15.5 26.3

CDSSTR Secondary structure assignment

α- helix (%) β-sheet (%) Turn (%) Unordered (%)

BLS 67.8 4.5 9 19.1

BLS refolded 77.2 2 5.9 15

M-GP5-BLS 54.9 9.7 12.2 23

SELCON3 Secondary structure assignment

α- helix (%) β-sheet (%) Turn (%) Unordered (%)

BLS 67.7 4.1 9.1 19.4

BLS refolded 71.2 3.3 7.7 19.1

M-GP5-BLS 53 7.9 15.1 25.2

49

3.8 Transmission electron microscopy shows pentameric assembly of both BLS and M-GP5-BLS

Nanoparticle assembly was visualized by transmission electron microscopy (TEM) for

both BLS and M-GP5-BLS. The two proteins, purified by size-exclusion

chromatography, were negatively stained with uranyl acetate and examined. Both BLS

and M-GP5-BLS appear to form pentameric structures with an approximate diameter of

5-7 nm (Figure 17). TEM images showed no noticeable difference in sizes between BLS

and M-GP5-BLS decamers. However, images obtained for BLS generally contained less

clusters than those obtained for M-GP5-BLS indicating that BLS is less likely to form

unwanted aggregates.

50

Figure 17. Negative stain electron microscopy analysis of BLS (A.) and M-GP5-BLS

(B.) purified by size-exclusion chromatography. Pentameric structures with diameters

between 5-7 nm were observed for both BLS and M-GP5-BLS. Scale bar: 50 nm.

51

Chapter 4

4 Discussion

To create a vaccine candidate for porcine reproductive and respiratory syndrome (PRRS),

four protein nanoparticles were screened as potential carriers for antigens derived from

the PRRS virus (PRRSV). These engineered nanoparticles were screened to discover

ones that were well overexpressed, soluble, stable and easily purified. Here, I report the

generation, purification and characterization of a PRRS vaccine candidate comprising a

fusion between Brucella lumazine synthase and select peptides from both major

structural proteins of PRRSV, the M protein (M) and glycoprotein 5 (GP5).

4.1 Selection of candidate PRRSV vaccine

Initially, all potential carriers (BLS, AaLS, Ferritin, sHSP-16.5) were genetically fused

with a GP5 peptide derived from the N-terminal ectodomain of the PRRSV GP5

membrane protein. In theory, the ectodomain of GP5 is an ideal peptide to display on the

nanoparticles of interest as it is conserved and immunoreactive (Mardassi et al., 1996;

Ostrowski et al., 2002; Plagemann et al., 2002; Wissink et al., 2003). I chose to display

residues 30 to 54 of GP5, encompassing the neutralizing epitopes identified

previously(Ostrowski et al., 2002; Plagemann et al., 2002; Wissink et al., 2003).

These fusion proteins expressed well in E. coli BL-21 cells when induced, however, they

aggregated and formed inclusion bodies which hindered further purification by IMAC.

Inclusion bodies (IBs) are dense, insoluble protein aggregates that are commonly

observed in E. coli cells expressing recombinant proteins (Baneyx and Mujacic, 2004).

IBs are generally considered to be a negative aspect of recombinant protein production

because proteins concentrated within IBs require additional processing such as

solubilization using a protein denaturant, and subsequent protein refolding steps. A

protein’s solubility has long been considered a major indicator of it having reached its

proper conformation. Thus, in this case where the fused nanoparticles are all almost

completely insoluble (Figure 8), it appears that the foreign peptide is preventing them

from folding properly to reach their usually soluble conformation.

52

The GP5 peptide I selected is composed of residues 30-54 of the GP5 membrane protein

from PRRSV and this peptide was selected for display on the nanoparticles of interest

because the sequence is exposed on the surface of the virion, well conserved and

immunogenic (Ostrowski et al., 2002; Plagemann et al., 2002). However, this sequence

includes four potential N-glycosylation sites at positions 30, 33, 44 and 51. In general, N-

glycosylation is important for proper folding and solubility of proteins and N-glycans

also play a role in the protein’s biological activity by serving as a recognition tag that

allows these proteins to interact with receptors involved in transmembrane signaling

(Helenius and Aebi, 2004). With respect to GP5, previous work showed that the potential

N-glycosylation sites N33, N44 and N51 are all occupied by glycan moieties in fully

matured PRRSV GP5 (Ansari et al., 2006). The glycosylation site N44 has been

repeatedly shown to be essential for viral particle formation; mutations at this site

drastically reduce formation, release and infectivity of PRRSV (Meulenberg et al., 1995;

Wissink et al., 2005). Since I used E. coli as an expression system, those N-glycans are

absent from the fusion proteins produced and it is likely that their absence is negatively

affecting the solubility of the PRRSV GP5 sequence chosen for display. Also in respect

to the absence of N-glycans present on the candidate vaccines, their efficacy is a concern

since in the case where neutralizing antibodies are produced against the chosen

ectodomain sequence, there is still the possibility that N-glycans present on the native

GP5 membrane protein will prevent binding of these antibodies to that ectodomain

sequence on the surface of the PRRS virion (Vu et al., 2011).

To overcome the insolubility of the engineered proteins and acquire soluble protein for

purification and future characterization, urea, a strong denaturant, was included in the

lysis buffer preparation at a concentration of 8M (Figure 9). The use of urea is acceptable

in laboratory-scale production for research purposes, however, from the standpoint of

industrial production, use of urea in the production of a vaccine would be greatly

discouraged. Urea essentially denatures proteins that form IBs and upon purification of

the desired recombinants proteins, refolding steps must be performed to recover

structured, functional recombinants. A significant portion of the GP5 engineered proteins

are lost during the refolding steps (Figure 11) and therefore proteins produced in IBs are

not marketable products; recovery is low, and the functionality of the product is

53

questioned. It has been common practice to discard recombinant proteins produced in IBs

(Baneyx and Mujacic, 2004). However, there are some recovery and refolding strategies

for proteins solubilized by urea but they are often only successful on a case by case basis

so there is still no standard for refolding and recovery of potential therapeutic proteins

from IBs (Ferrer-Miralles et al., 2009). To recover the recombinant proteins from their

denatured state in urea, I first performed stepwise dialysis in 2M increments from 8M

urea to no urea, but acquired similar recovery levels as with direct dialysis into 50 mM

Tris-HCl (pH 8.5), 100 mM NaCl, 14.3 mM 2-mercaptoethanol, so the latter was kept as

routine.

I was able to enrich for the engineered proteins by Ni-NTA chromatography under

denaturing conditions (Figure 10B) and later dialyze eluates to recover soluble

engineered proteins (Figure 11). Although my objective is to have one of these

nanoparticles serve as a platform for the PRRSV antigen, it would be preferable that the

antigen be accepted without disrupting the native folding of the core structure of the

nanoparticles. Initially, I sought to achieve this by genetically fusing the chosen antigen

to the solvent exposed N-termini (BLS, Ferritin) or C-termini (AaLS, HSP) of our

nanoparticles of interest. However, it was evident by the amount of aggregation that

folding was disrupted and therefore, at this point, I investigated options to minimize

aggregation. General approaches to minimize recombinant protein aggregation in E. coli

include chaperone protein co-expression, temperature and transcription control, and

protein engineering (Garcia-Fruitos et al., 2012). As a first approach, expression of all

engineered proteins was induced at 16°C overnight instead of the usual 37°C for 3 hours,

however, there was no apparent increase in the yield of soluble protein. Next, two genetic

constructs were designed using BLS as the platform; M-GP5-BLS and GP5-M-BLS.

Briefly, these constructs contain the first ectodomain of the PRRSV M protein, linked to

my chosen GP5 antigen by a flexible linker, and finally this M-GP5 or GP5-M sequence,

is linked genetically to the N-terminus of BLS via a flexible linker (Figure 12).

The decision to produce an antigen with both M and GP5 ectodomain sequences was

made based on past research indicating that the GP5-M heterodimers that dominate the

envelope membrane of PRRSV are linked via a disulfide bond between their ectodomains

54

(de Vries et al., 1995; Faaberg et al., 1995; Mardassi et al., 1996). Apart from increasing

the yield of the soluble proteins, it is possible that with this disulfide bond within my

vaccine candidates, I may present epitopes that are specific to the GP5-M heterodimer

that could not be targeted with the display of the GP5 ectodomain alone. Both

participating cysteines are included in the constructs. However, since I used E. coli as the

expression system, this interaction will not occur upon production of the fusion proteins

since, the reducing environment maintained in E. coli cells will prevent disulfide bond

formation. Therefore, I hypothesized that the pairing of M and GP5 on the nanoparticles

may help minimize aggregation at the refolding stage in the presence of non-reducing

environment where disulfide bond formation can readily occur and potentially assist in

folding of the fusion proteins.

Expression plasmids containing the new constructs, either M-GP5-BLS and GP5-M-BLS,

were transformed into E. coli BL-21 cells, and expressed and purified by IMAC. M-GP5-

BLS appeared to overexpress very well in E. coli while GP5-M-BLS did not (Figure

13A). As with the previous fusion proteins, which contained the GP5 antigen alone, I

found that both M-GP5-BLS and GP5-M-BLS fusion proteins formed inclusion bodies in

E. coli (Figure 13B) and thus, urea was needed for their extraction. Upon nickel

purification of M-GP5-BLS, it was acquired with a high level of purity with one

prominent protein species at its expected size on the SDS-PAGE gel. However, upon

purification of GP5-M-BLS, there were three prominent protein species with the highest

at the expected size of GP5-M-BLS and two lower molecular weight species (Figure

13C). These lower molecular weight species are thought to be degradation products of

GP5-M-BLS since they do not appear on the SDS-PAGE gel of M-GP5-BLS nickel

purification (Figure 13C). Both GP5-M-BLS and M-GP5-BLS were cloned into identical

expression plasmids (pET-28a+) and transformed into identical E coli strains. If these

lower molecular weight protein species were not associated with expression of the GP5-

M-BLS fusion protein, they would be purified in a similar fashion from the E. coli cells

transformed with the plasmid containing M-GP5-BLS. Therefore, I concluded that the

fusion protein GP5-M-BLS was prone to degradation and thus, the order in which the

selected M and GP5 sequences were fused to BLS affected its stability. Although

55

insoluble, the fusion protein M-GP5-BLS expressed well in E. coli, was stable and could

be significantly enriched via Ni-NTA chromatography.

To investigate whether the presence of the ectodomain of the PRRSV M protein helped

with the refolding of the new fusion protein M-GP5-BLS, the nickel purified sample was

dialyzed exhaustively against Tris buffer containing no denaturant with either reducing

agent or no reducing agent (Figure 13D). Supernatant and pelleted fractions of both

dialyses did not indicate any advantage of dialyzing without reducing agent so there was

no indication that interaction between M and GP5 via a disulfide bridge helped in the

refolding of this fusion protein. Comparing the fractions of each dialysis separately, I saw

that without reducing agent, a large percentage of M-GP5-BLS aggregated while with

reducing agent, a significantly smaller percentage of M-GP5-BLS aggregated (Figure

13D). Therefore, it appeared that the presence of reducing agent helped M-GP5-BLS

retain solubility upon removal of urea, as it did with the fused nanoparticles previously

tested and other recombinantly expressed fusion proteins of BLS (Bellido et al., 2009).

My best candidate thus far is M-GP5-BLS; it overexpressed well in E. coli, was stable,

and most of it remained soluble upon removal of urea following Ni-NTA enrichment.

4.2 Structural characterization of BLS and M-GP5-BLS

Both BLS and M-GP5-BLS were purified further and analyzed by size-exclusion

chromatography (SEC). The elution profile of the recombinantly-produced BLS showed

a single, narrow, monodispersed peak eluting at an apparent molecular weight slightly

higher than 158 kDa (Figure 14C). Since this peak when analyzed by SDS-PAGE

contained the pure monomer of BLS which has a molecular weight of 18.8 kDa, it is

evident that monomers of BLS assembled into the expected decameric form with a

molecular weight of 188 kDa. The molecular weight of BLS was estimated to be 172 kDa

based on the linear calibration curve created by plotting the elution volumes of the

protein standards used to calibrate the column versus the log10 of their molecular weights.

Based on the elution profile of M-GP5-BLS, it is evident that this fusion protein

assembles into a multimeric complex. However, its elution profile shows a peak at a

56

much earlier volume (11.6 ml) than both the peak of 158 kDa protein marker (13.11 ml)

and that of BLS (13 ml). Based on the SEC calibration curve, M-GP5-BLS is estimated at

a size of 505 kDa. Estimating the molecular weight of a protein using SEC is

recommended only when the protein takes a shape similar to the proteins used to calibrate

the column; in this case globular proteins were used for calibration. This is recommended

because SEC elution profiles are affected by both the size and shape of molecule. Two

molecules with identical or very similar molecular weights will not have the same elution

volume if one has a rod-like shape and the other is a compact sphere (Sorensen et al.,

2001). The rod-shaped molecule will pass through the SEC column at a faster rate

because it has less access to the small pores within the resin of the column than a

molecule that is compact and globular. BLS is cylindrical in shape, referred to as spool-

like (Zylberman et al., 2004) but it is still a small compact molecule and this is likely why

its size estimation was close to its calculated molecular weight; BLS does not appear to

depart much from a globular shape. Since M-GP5 is predicted as mainly disordered

(Kelley et al., 2015) it is likely that it extends randomly from the BLS core and causes a

great departure from the compact spool-like shape of the BLS molecule. This change in

shape to a more elongated molecule would explain the surprisingly large shift in elution

volume between BLS and M-GP5-BLS, and the potentially over-estimated molecular

weight of 505 kDa.

To acquire more precise estimates of the shape and molecular weights (MWs) for BLS

and M-GP5-BLS sedimentation velocity (SV) AUC was performed. Along with MW

estimation, SV experiments also estimate the frictional ratio of a sample and this value

gives a clear indication of the shape of a molecule in solution. Results from my SV

experiment for BLS (Figure 15A) agreed with the SEC data, wherein BLS has a singular,

monodispersed size distribution with an estimated molecular weight of 178 kDa. As for

the frictional ratio of BLS, it was estimated to be 1.3. By definition, the frictional ratio is

a measure of the resistance experienced by a molecule during sedimentation in relation to

the resistance experienced by an ideal sphere of the same molecular weight (Smith,

1988). Therefore, an ideal sphere of any molecular weight will theoretically have a

frictional ratio of 1.0 and if this sphere shifts towards a more elongated form, it will

experience more resistance and an increase in frictional ratio. BLS therefore departs from

57

the ideal sphere and is more elongated but it is still considered to be globular since

globular proteins typically have frictional ratios ranging from 1.05 to 1.30 (Unzai, 2018).

The globular nature of BLS explains the relatively accurate estimation of its MW by

SEC. Based on the SV experiments of M-GP5-BLS, the engineered construct sediments

more quickly than BLS, having a sedimentation coefficient of 8.8 compared to the 8.1

calculated for BLS. Multiple peaks in the sedimentation plot of M-GP5-BLS (Figure

15B) indicated that the engineered construct formed aggregates. Estimation of molecular

mass by SV experiments is accurate and considered most appropriate when analyzing

heterogenous samples with molecules having similar frictional ratios or when a sample

has a single major peak in its c(s) distribution plot (Dam and Schuck, 2004) as seen with

BLS (Figure 15A). In the case of M-GP5-BLS, since there is aggregation and the

appearance of multiple species, a weight-average frictional ratio is calculated to represent

the frictional ratio of all species therein. It was calculated to be 1.58, indicative of an

elongated molecule no longer considered as globular. Although the weighted-average

frictional ratio represents the frictional ratio of the most abundant species in a sample

well, the molecular mass estimate is not expected to be as accurate. The estimated

molecular mass of M-GP5-BLS by SV is 220 kDa, which is 32 kDa less than the

expected MW of 252 kDa, however, within the experimental uncertainty, this is

consistent with a decamer of M-GP5-BLS.

In order to evaluate and compare the secondary structure of BLS and M-GP5-BLS, their

circular dichroism (CD) signal in the far UV (190–260 nm wavelength range) was

measured. The far UV-circular dichroism spectra of BLS was practically superimposable

with that of the BLS which was extracted with urea and later refolded (Figure 16A). This

shows us that refolding by dialysis allowed for the proper folding of BLS with no

disruption of its native secondary structure and therefore it is inferred that refolding of M-

GP5-BLS follows this trend and so its BLS core structure folds properly. Although the

spectra of BLS and M-GP5-BLS do not overlay (Figure 16A), they do have an identical

overall shape and a shared minimum of 220 nm, indicating that the overall secondary

structure of BLS is not changed by the addition of M-GP5. Both spectra correspond to a

protein with majority alpha helical structure and some beta secondary structure, which

agrees with the crystallographic structure of BLS (Zylberman et al., 2004).

58

Analysis of the thermal stability of BLS and M-GP5-BLS by CD spectroscopy and the

transmission electron microscopy (TEM) images taken of these two recombinant proteins

indicate that the presence of the PRRSV antigenic sequence M-GP5 does not disrupt the

overall structure of BLS. Both BLS and M-GP5-BLS share similar melting temperatures

(Tms), 92.1°C and 91.5°C respectively (Figures 15B and 15C), and their high Tms are

typical of a protein from a thermophilic organism. The thermal stability of these two

recombinant proteins also speaks to their oligomeric state. Based on the reported Tms of

lumazine synthases with pentameric or icosahedral (60 subunits) assemblies, my

decameric BLS is expected to have an intermediate Tm. Zhang and colleagues (2001)

reported the melting temperature of the pentameric lumazine synthase isolated from

Saccharomyces cerevisiae as 74.1°C and the melting temperature of the icosahedral

lumazine synthase isolated from Aquifex aeolicus as 119.9°C. The Tm of BLS I have

estimated based on my CD experiments is 92.1°C; it lies between 74.1°C and 119.9°C

and is very similar to the BLS Tm of 88 ± 2°C estimated by the CD experiments of

Zylberman and colleagues (2004). Therefore, it is clear that the M-GP5-BLS chimera

produced in this work maintains the thermal stability of BLS and likely assembles into a

decamer similar to BLS. Negative stain TEM images show that both BLS and M-GP5-

BLS appear to form pentameric structures with an approximate diameter of 5-7 nm

(Figure 17). However, my results from SEC, SV and the thermal denaturation

experiments strongly indicate the formation of a higher order, more stable decameric

species, not the pentamer visualized by TEM. Looking from above the decameric barrel

of BLS, it has a diameter of 7 nm and from the side, a height of 8 nm (Zylberman et al.,

2004). Because of the similarity in size of these two major orientations, it is likely that in

the TEM images BLS and M-GP5-BLS are in several orientations, but the top view is

easiest to interpret. Also, it is possible that the low pH of the uranyl acetate (pH 4-4.5)

used to stain the samples caused the dissociation of decameric BLS and M-GP5-BLS into

their pentameric state. Zylberman and his colleague (2004) report dissociation of

decameric BLS into pentamers between the pH range 4.0-5.0 at room temperature.

Visualizing the engineered parts of M-GP5-BLS and thus distinguishing BLS from M-

GP5-BLS by TEM was not possible; likely due to the disordered nature of the fused M-

GP5 sequence.

59

4.3 Conclusion and future directions

To conclude, I show that I have successfully produced a Brucella lumazine synthase

nanoparticle that carries antigenic peptide sequences from the ectodomains of both major

structural proteins of the PRRS virus; M protein and glycoprotein 5. The antigenic

sequences (M-GP5) were genetically fused as one foreign peptide to the N-terminus of

BLS and my results demonstrated that the secondary structure of BLS was conserved in

the chimeric protein M-GP5-BLS, the thermal stability of BLS was unaltered and that M-

GP5-BLS assembles into a decamer similar to the wild-type BLS.

The next step for this project will be to determine whether M-GP5-BLS induces an

immune response against PRRSV. This can be studied in mice by subcutaneous

injections of the recombinant protein followed by monitoring the production and level of

PRRSV-specific antibodies and neutralizing antibodies (Yu et al., 2016). Typically, a

laboratory-bred strain of mice called BALB/c mice would be the test subjects; this strain

of house mice is routinely used in animal experimentation and will lack exposure to

PRRSV. The test group would be immunized with my test vaccine, blood sera would be

collected, and antibody activity determined using indirect enzyme-linked immunosorbent

assay (ELISA) kits coated with PRRSV antigens. Sera containing PRRSV-specific

antibodies can then be tested to determine whether they can prevent the virus from

infecting susceptible cells. M-GP5-BLS is expected to stimulate neutralizing antibodies

against the immunoreactive domain of GP5 and since this is region is conserved across

PRRSV strains of the NA genotype (Figure 7), it is likely that these antibodies will bind

the GP5 ectodomain of various strains of the NA genotype. With respect to the divergent

EU genotype, protection may be elicited if antibodies target specifically the region that is

conserved between both genotypes (Figure 7). To improve on my potential vaccine, a

counterpart M-GP5-BLS could be made that is based on the M and GP5 membrane

proteins of PRRSV strains from the EU genotype. A mixture of these two M-GP5-BLS

constructs is likely to increase the potential of protection against a vast range of PRRSV

isolates.

60

Another direction that may be explored is the production of my engineered proteins in

plants. Transgenic Arabidopsis and tobacco plants are common expression systems in

which glycosylation of my constructs can be controlled via subcellular targeting (An et

al., 2018; Pereira et al., 2014) and with these glycosylated constructs purified, their

antigenicity can be tested and compared with that of the non-glycosylated constructs

produced in E. coli. Glycosylation may affect the antigenicity of our constructs since in

some cases, glycans are involved in the interaction between antibody and epitope and in

other cases, glycans shield epitopes from antibodies and protect viruses from the immune

system (Desrosiers et al., 2004; Lisowska, 2002; Vu et al., 2011). In the case where the

non-glycosylated constructs induce antibodies against an epitope shielded by a glycan on

the surface of native PRRSV, immunological testing will reveal the efficacy of those

constructs as vaccines. It’s possible that the shielded epitope is exposed along the course

of the virus’ infectious cycle and upon exposure of the epitope, antibodies may bind and

provide control of the virus.

In summary, I used BLS as a platform for PRRSV antigens to create a PRRS vaccine

candidate. Its efficacy must be investigated, and improvements will likely be necessary to

promote cross-protective immunity. Alternative methods of protein production will also

be explored to investigate the importance of glycosylation of the PRRSV antigens on the

antigenicity and efficacy of the vaccine candidate.

61

References

Alfano, E.F., Lentz, E.M., Bellido, D., Dus Santos, M.J., Goldbaum, F.A., Wigdorovitz, A., Bravo-

Almonacid, F.F., 2015. Expression of the Multimeric and Highly Immunogenic Brucella spp.

Lumazine Synthase Fused to Bovine Rotavirus VP8d as a Scaffold for Antigen Production in

Tobacco Chloroplasts . Frontiers in plant science . https://doi.org/10.3389/fpls.2015.01170

Allende, R., Lewis, T.L., Lu, Z., Rock, D.L., Kutish, G.F., Ali, A., Doster, A.R., Osorio, F.A., 1999. North

American and European porcine reproductive and respiratory syndrome viruses differ in non-

structural protein coding regions. The Journal of general virology 80 ( Pt 2), 307–315.

https://doi.org/10.1099/0022-1317-80-2-307

An, C.H., Nazki, S., Park, S.-C., Jeong, Y.J., Lee, J.H., Park, S.-J., Khatun, A., Kim, W.-I., Park, Y.-I.,

Jeong, J.C., Kim, C.Y., 2018. Plant synthetic GP4 and GP5 proteins from porcine reproductive and

respiratory syndrome virus elicit immune responses in pigs . Planta . https://doi.org/10.1007/s00425-

017-2836-z

Ansari, I.H., Kwon, B., Osorio, F.A., Pattnaik, A.K., 2006. Influence of N-linked glycosylation of porcine

reproductive and respiratory syndrome virus GP5 on virus infectivity, antigenicity, and ability to

induce neutralizing antibodies. Journal of virology 80, 3994–4004.

https://doi.org/10.1128/JVI.80.8.3994-4004.2006

Ashley, C.E., Carnes, E.C., Phillips, G.K., Durfee, P.N., Buley, M.D., Lino, C.A., Padilla, D.P., Phillips,

B., Carter, M.B., Willman, C.L., Brinker, C.J., Caldeira, J. do C., Chackerian, B., Wharton, W.,

Peabody, D.S., 2011. Cell-specific delivery of diverse cargos by bacteriophage MS2 virus-like

particles. ACS nano 5, 5729–5745. https://doi.org/10.1021/nn201397z

Baneyx, F., Mujacic, M., 2004. Recombinant protein folding and misfolding in Escherichia coli. Nature

biotechnology 22, 1399–1408. https://doi.org/10.1038/nbt1029

Bellido, D., Craig, P.O., Mozgovoj, M. V, Gonzalez, D.D., Wigdorovitz, A., Goldbaum, F.A., Dus Santos,

M.J., 2009. Brucella spp. lumazine synthase as a bovine rotavirus antigen delivery system. Vaccine

27, 136–145. https://doi.org/10.1016/j.vaccine.2008.10.018

Benfield, D.A., Nelson, E., Collins, J.E., Harris, L., Goyal, S.M., Robison, D., Christianson, W.T.,

Morrison, R.B., Gorcyca, D., Chladek, D., 1992. Characterization of swine infertility and respiratory

syndrome (SIRS) virus (isolate ATCC VR-2332). Journal of veterinary diagnostic investigation :

official publication of the American Association of Veterinary Laboratory Diagnosticians, Inc 4,

127–133. https://doi.org/10.1177/104063879200400202

62

Botner, A., Strandbygaard, B., Sorensen, K.J., Have, P., Madsen, K.G., Madsen, E.S., Alexandersen, S.,

1997. Appearance of acute PRRS-like symptoms in sow herds after vaccination with a modified live

PRRS vaccine. The Veterinary record 141, 497–499. https://doi.org/10.1136/vr.141.19.497

Bova, M.P., Huang, Q., Ding, L., Horwitz, J., 2002. Subunit Exchange, Conformational Stability, and

Chaperone-like Function of the Small Heat Shock Protein 16.5 fromMethanococcus jannaschii .

Journal of Biological Chemistry 277, 38468–38475. https://doi.org/10.1074/jbc.M205594200

Braden, B.C., Velikovsky, C.A., Cauerhff, A.A., Polikarpov, I., Goldbaum, F.A., 2000. Divergence in

macromolecular assembly: X-ray crystallographic structure analysis of lumazine synthase from

Brucella abortus. Journal of molecular biology 297, 1031–1036.

https://doi.org/10.1006/jmbi.2000.3640

Bundy, B.C., Franciszkowicz, M.J., Swartz, J.R., 2008. Escherichia coli-based cell-free synthesis of virus-

like particles. Biotechnology and bioengineering 100, 28–37. https://doi.org/10.1002/bit.21716

CAN, L., YI-BAO, N., BIN-RUI, X., WEN-ZHI, G., DONG-DONG, Z., 2016. Analysis of genetic

variation of porcine reproductive and respiratory syndrome virus (PRRSV) isolates in Central China .

Journal of Veterinary Medical Science . https://doi.org/10.1292/jvms.15-0570

Cassataro, J., Pasquevich, K.A., Estein, S.M., Laplagne, D.A., Velikovsky, C.A., de la Barrera, S., Bowden,

R., Fossati, C.A., Giambartolomei, G.H., Goldbaum, F.A., 2007. A recombinant subunit vaccine

based on the insertion of 27 amino acids from Omp31 to the N-terminus of BLS induced a similar

degree of protection against B. ovis than Rev.1 vaccination. Vaccine 25, 4437–4446.

https://doi.org/10.1016/j.vaccine.2007.03.028

Cervera, L., Gutierrez-Granados, S., Martinez, M., Blanco, J., Godia, F., Segura, M.M., 2013. Generation

of HIV-1 Gag VLPs by transient transfection of HEK 293 suspension cell cultures using an optimized

animal-derived component free medium. Journal of biotechnology 166, 152–165.

https://doi.org/10.1016/j.jbiotec.2013.05.001

Chaplin, D.D., 2010. Overview of the immune response . J Allergy Clin Immunol .

https://doi.org/10.1016/j.jaci.2009.12.980

Chen, J., Liu, T., Zhu, C.-G., Jin, Y.-F., Zhang, Y.-Z., 2006. Genetic Variation of Chinese PRRSV Strains

Based on ORF5 Sequence . Biochemical Genetics . https://doi.org/10.1007/s10528-006-9039-9

Cho, K.J., Shin, H.J., Lee, C., Lee, Y., Lee, J.-H., Kim, K.H., Kim, K.-J., Park, Sarah S, Park, Sung Soo,

2009. The Crystal Structure of Ferritin from Helicobacter pylori Reveals Unusual Conformational

Changes for Iron Uptake . Journal of Molecular Biology . https://doi.org/10.1016/j.jmb.2009.04.078

63

Christianson, W.T., 1992. Stillbirths, mummies, abortions, and early embryonic death. The Veterinary

clinics of North America. Food animal practice 8, 623–639.

Chua, B.Y., Pejoski, D., Turner, S.J., Zeng, W., Jackson, D.C., 2011. Soluble Proteins Induce Strong

CD8&lt;sup&gt;+&lt;/sup&gt; T Cell and Antibody Responses through Electrostatic Association

with Simple Cationic or Anionic Lipopeptides That Target TLR2. The Journal of Immunology 187,

1692 LP – 1701. https://doi.org/10.4049/jimmunol.1100486

Crooks, G.E., Hon, G., Chandonia, J.-M., Brenner, S.E., 2004. WebLogo: A sequence logo generator .

Genome Research . https://doi.org/10.1101/gr.849004

Dalmau, M., Lim, S., Chen, H.C., Ruiz, C., Wang, S., 2008. Thermostability and molecular encapsulation

within an engineered caged protein scaffold . Biotechnology and Bioengineering .

https://doi.org/10.1002/bit.21988

Dam, J., Schuck, P., 2004. Calculating sedimentation coefficient distributions by direct modeling of

sedimentation velocity concentration profiles. Methods in enzymology 384, 185–212.

https://doi.org/10.1016/S0076-6879(04)84012-6

de Lima, M., Pattnaik, A.K., Flores, E.F., Osorio, F.A., 2006. Serologic marker candidates identified

among B-cell linear epitopes of Nsp2 and structural proteins of a North American strain of porcine

reproductive and respiratory syndrome virus. Virology 353, 410–421.

https://doi.org/10.1016/j.virol.2006.05.036

de Vries, A.A., Post, S.M., Raamsman, M.J., Horzinek, M.C., Rottier, P.J., 1995. The two major envelope

proteins of equine arteritis virus associate into disulfide-linked heterodimers. Journal of virology 69,

4668–4674.

Dea, S., Gagnon, C.A., Mardassi, H., Pirzadeh, B., Rogan, D., 2000. Current knowledge on the structural

proteins of porcine reproductive and respiratory syndrome (PRRS) virus: comparison of the North

American and European isolates. Archives of virology 145, 659–688.

Desrosiers, R.C., Doms, R.W., Sodroski, J., Moore, J.P., Wyatt, R.T., Nabel, G.J., Wilson, I.A., Burton,

D.R., Kwong, P.D., Koff, W.C., 2004. HIV vaccine design and the neutralizing antibody problem .

Nature Immunology . https://doi.org/10.1038/ni0304-233

Domingo, G.J., Orru’, S., Perham, R.N., 2001. Multiple Display of Peptides and Proteins on a

Macromolecular Scaffold Derived from a Multienzyme Complex . Journal of Molecular Biology .

https://doi.org/10.1006/jmbi.2000.4311

64

Faaberg, K.S., Even, C., Palmer, G.A., Plagemann, P.G., 1995. Disulfide bonds between two envelope

proteins of lactate dehydrogenase-elevating virus are essential for viral infectivity. Journal of

virology 69, 613–617.

Ferrer-Miralles, N., Domingo-Espin, J., Corchero, J.L., Vazquez, E., Villaverde, A., 2009. Microbial

factories for recombinant pharmaceuticals. Microbial cell factories 8, 17.

https://doi.org/10.1186/1475-2859-8-17

Flenniken, M.L., Liepold, L.O., Crowley, B.E., Willits, D.A., Young, M.J., Douglas, T., 2005. Selective

attachment and release of a chemotherapeutic agent from the interior of a protein cage architecture.

Chemical communications (Cambridge, England) 447–449. https://doi.org/10.1039/b413435d

Flenniken, M.L., Willits, D.A., Brumfield, S., Young, M.J., Douglas, T., 2003. The Small Heat Shock

Protein Cage from Methanococcus jannaschii Is a Versatile Nanoscale Platform for Genetic and

Chemical Modification. Nano Letters 3, 1573–1576. https://doi.org/10.1021/nl034786l

Flenniken, M.L., Willits, D.A., Harmsen, A.L., Liepold, L.O., Harmsen, A.G., Young, M.J., Douglas, T.,

2006. Melanoma and Lymphocyte Cell-Specific Targeting Incorporated into a Heat Shock Protein

Cage Architecture. Chemistry & Biology 13, 161–170.

https://doi.org/10.1016/j.chembiol.2005.11.007

Frietze, K.M., Peabody, D.S., Chackerian, B., 2016. Engineering virus-like particles as vaccine platforms.

Current Opinion in Virology 18, 44–49. https://doi.org/https://doi.org/10.1016/j.coviro.2016.03.001

Garcia-Fruitos, E., Vazquez, E., Diez-Gil, C., Corchero, J.L., Seras-Franzoso, J., Ratera, I., Veciana, J.,

Villaverde, A., 2012. Bacterial inclusion bodies: making gold from waste. Trends in biotechnology

30, 65–70. https://doi.org/10.1016/j.tibtech.2011.09.003

Heddle, J.G., Chakraborti, S., Iwasaki, K., 2017. Natural and artificial protein cages: design, structure and

therapeutic applications. Current opinion in structural biology 43, 148–155.

https://doi.org/10.1016/j.sbi.2017.03.007

Helenius, A., Aebi, M., 2004. Roles of N-linked glycans in the endoplasmic reticulum. Annual review of

biochemistry 73, 1019–1049. https://doi.org/10.1146/annurev.biochem.73.011303.073752

Holtkamp, D.J., Kliebenstein, J.B., Neumann, E.J., 2013. Assessment of the economic impact of porcine

reproductive and respiratory syndrome virus on United States pork producers. Journal of Swine

Health and Production 21, 72–84.

Hsia, Y., Bale, J.B., Gonen, S., Shi, D., Sheffler, W., Fong, K.K., Nattermann, U., Xu, C., Huang, P.-S.,

65

Ravichandran, R., Yi, S., Davis, T.N., Gonen, T., King, N.P., Baker, D., 2016. Design of a

hyperstable 60-subunit protein icosahedron . Nature . https://doi.org/10.1038/nature18010

Jameson, G.N.L., Jin, W., Krebs, C., Perreira, A.S., Tavares, P., Liu, X., Theil, E.C., Huynh, B.H., 2002.

Stoichiometric production of hydrogen peroxide and parallel formation of ferric multimers through

decay of the diferric-peroxo complex, the first detectable intermediate in ferritin mineralization.

Biochemistry 41, 13435–13443. https://doi.org/10.1021/bi026478s

Jiang, W., Jiang, P., Li, Y., Tang, J., Wang, X., Ma, S., 2006. Recombinant adenovirus expressing GP5 and

M fusion proteins of porcine reproductive and respiratory syndrome virus induce both humoral and

cell-mediated immune responses in mice. Veterinary immunology and immunopathology 113, 169–

180. https://doi.org/10.1016/j.vetimm.2006.05.001

Johnson, C., 2012. Research to Make Alberta PRRS-Free [WWW Document]. The Pig Site. URL

https://thepigsite.com/news/2012/09/research-to-make-alberta-prrsfree (accessed 7.24.19).

Kabsch, Wolfgang, Kabsch, W, Sander, C, Sander, Christian, 1983. Dictionary of protein secondary

structure: Pattern recognition of hydrogen‐bonded and geometrical features . Biopolymers .

https://doi.org/10.1002/bip.360221211

Kanekiyo, M., Wei, C.-J., Yassine, H.M., McTamney, P.M., Boyington, J.C., Whittle, J.R.R., Rao, S.S.,

Kong, W.-P., Wang, L., Nabel, G.J., 2013. Self-assembling influenza nanoparticle vaccines elicit

broadly neutralizing H1N1 antibodies. Nature 499, 102+.

Kang, H.J., Kang, Y.J., Lee, Y.-M., Shin, H.-H., Chung, S.J., Kang, S., 2012. Developing an antibody-

binding protein cage as a molecular recognition drug modular nanoplatform. Biomaterials 33, 5423–

5430. https://doi.org/10.1016/j.biomaterials.2012.03.055

Kang, Y.J., Park, D.C., Shin, H.-H., Park, J., Kang, S., 2012. Incorporation of thrombin cleavage peptide

into a protein cage for constructing a protease-responsive multifunctional delivery nanoplatform.

Biomacromolecules 13, 4057–4064. https://doi.org/10.1021/bm301339s

Kelley, L.A., Mezulis, S., Yates, C.M., Wass, M.N., Sternberg, M.J.E., 2015. The Phyre2 web portal for

protein modeling, prediction and analysis. Nature Protocols 10, 845.

Kim, M., Rho, Y., Jin, K.S., Ahn, B., Jung, S., Kim, H., Ree, M., 2011. pH-dependent structures of ferritin

and apoferritin in solution: disassembly and reassembly. Biomacromolecules 12, 1629–1640.

https://doi.org/10.1021/bm200026v

Klinke, S., Zylberman, V., Vega, D.R., Guimaraes, B.G., Braden, B.C., Goldbaum, F.A., 2005.

66

Crystallographic studies on decameric Brucella spp. Lumazine synthase: a novel quaternary

arrangement evolved for a new function? Journal of molecular biology 353, 124–137.

https://doi.org/10.1016/j.jmb.2005.08.017

Lager, K.M., Mengeling, W.L., Brockmeier, S.L., 1999. Evaluation of protective immunity in gilts

inoculated with the NADC-8 isolate of porcine reproductive and respiratory syndrome virus

(PRRSV) and challenge-exposed with an antigenically distinct PRRSV isolate. American journal of

veterinary research 60, 1022–1027.

Laplagne, D.A., Zylberman, V., Ainciart, N., Steward, M.W., Sciutto, E., Fossati, C.A., Goldbaum, F.A.,

2004. Engineering of a polymeric bacterial protein as a scaffold for the multiple display of peptides.

Proteins 57, 820–828. https://doi.org/10.1002/prot.20248

Lilavivat, S., Sardar, D., Jana, S., Thomas, G.C., Woycechowsky, K.J., 2012. In vivo encapsulation of

nucleic acids using an engineered nonviral protein capsid. Journal of the American Chemical Society

134, 13152–13155. https://doi.org/10.1021/ja302743g

Linder, M.C., Kakavandi, H.R., Miller, P., Wirth, P.L., Nagel, G.M., 1989. Dissociation of ferritins.

Archives of biochemistry and biophysics 269, 485–496. https://doi.org/10.1016/0003-

9861(89)90132-x

Lisowska, E., 2002. The role of glycosylation in protein antigenic properties . Cellular and Molecular Life

Sciences . https://doi.org/10.1007/s00018-002-8437-3

Liu, Y., Ye, L., Lin, F., Gomaa, Y., Flyer, D., Carrion, R., Patterson, J.L., Prausnitz, M.R., Smith, G.,

Glenn, G., Wu, H., Compans, R.W., Yang, C., 2018. Intradermal Vaccination With Adjuvanted Ebola

Virus Soluble Glycoprotein Subunit Vaccine by Microneedle Patches Protects Mice Against Lethal

Ebola Virus Challenge. The Journal of Infectious Diseases 218, S545–S552.

https://doi.org/10.1093/infdis/jiy267

López-Sagaseta, J., Malito, E., Rappuoli, R., Bottomley, M.J., 2016. Self-assembling protein nanoparticles

in the design of vaccines . Computational and Structural Biotechnology Journal .

https://doi.org/10.1016/j.csbj.2015.11.001

Mardassi, H., Massie, B., Dea, S., 1996. Intracellular synthesis, processing, and transport of proteins

encoded by ORFs 5 to 7 of porcine reproductive and respiratory syndrome virus. Virology 221, 98–

112. https://doi.org/10.1006/viro.1996.0356

Mardassi, H., Mounir, S., Dea, S., 1995. Molecular analysis of the ORFs 3 to 7 of porcine reproductive and

respiratory syndrome virus, Quebec reference strain. Archives of virology 140, 1405–1418.

67

Martelli, P., Gozio, S., Ferrari, L., Rosina, S., De Angelis, E., Quintavalla, C., Bottarelli, E., Borghetti, P.,

2009. Efficacy of a modified live porcine reproductive and respiratory syndrome virus (PRRSV)

vaccine in pigs naturally exposed to a heterologous European (Italian cluster) field strain: Clinical

protection and cell-mediated immunity . Vaccine . https://doi.org/10.1016/j.vaccine.2009.03.028

McAleer, W.J., Miller, W.J., Buynak, E.B., Maigetter, R.Z., Wampler, D.E., Hilleman, M.R., 1984. Human

hepatitis B vaccine from recombinant yeast. Nature 307, 178–180. https://doi.org/10.1038/307178a0

Mengeling, W.L., Lager, K.M., Vorwald, A.C., Koehler, K.J., 2003. Strain specificity of the immune

response of pigs following vaccination with various strains of porcine reproductive and respiratory

syndrome virus . Veterinary Microbiology . https://doi.org/10.1016/S0378-1135(02)00427-3

Meulenberg, J.J., Petersen-den Besten, A., De Kluyver, E.P., Moormann, R.J., Schaaper, W.M.,

Wensvoort, G., 1995. Characterization of proteins encoded by ORFs 2 to 7 of Lelystad virus.

Virology 206, 155–163. https://doi.org/10.1016/s0042-6822(95)80030-1

Min, J., Kim, S., Lee, J., Kang, S., 2014. Lumazine synthase protein cage nanoparticles as modular delivery

platforms for targeted drug delivery. RSC Advances 4, 48596–48600.

https://doi.org/10.1039/C4RA10187A

Miyanohara, A., Imamura, T., Araki, M., Sugawara, K., Ohtomo, N., Matsubara, K., 1986. Expression of

hepatitis B virus core antigen gene in Saccharomyces cerevisiae: synthesis of two polypeptides

translated from different initiation codons. Journal of Virology 59, 176–180.

Murtaugh, M.P., Stadejek, T., Abrahante, J.E., Lam, T.T.Y., Leung, F.C.-C., 2010. The ever-expanding

diversity of porcine reproductive and respiratory syndrome virus. Virus research 154, 18–30.

https://doi.org/10.1016/j.virusres.2010.08.015

Nathues, C., Perler, L., Bruhn, S., Suter, D., Eichhorn, L., Hofmann, M., Nathues, H., Baechlein, C.,

Ritzmann, M., Palzer, A., Grossmann, K., Schüpbach-Regula, G., Thür, B., 2016. An Outbreak of

Porcine Reproductive and Respiratory Syndrome Virus in Switzerland Following Import of Boar

Semen. Transboundary and Emerging Diseases 63, e251–e261. https://doi.org/10.1111/tbed.12262

Nelsen, C.J., Murtaugh, M.P., Faaberg, K.S., 1999. Porcine Reproductive and Respiratory Syndrome Virus

Comparison: Divergent Evolution on Two Continents . Journal of Virology .

Neumann, E.J., Kliebenstein, J.B., Johnson, C.D., Mabry, J.W., Bush, E.J., Seitzinger, A.H., Green, A.L.,

Zimmerman, J.J., 2005. Assessment of the economic impact of porcine reproductive and respiratory

syndrome on swine production in the United States. Journal of the American Veterinary Medical

Association 227, 385–392.

68

Noad, R., Roy, P., 2003. Virus-like particles as immunogens . Trends in Microbiology .

https://doi.org/10.1016/S0966-842X(03)00208-7

Ostrowski, M., Galeota, J.A., Jar, A.M., Platt, K.B., Osorio, F.A., Lopez, O.J., 2002. Identification of

neutralizing and nonneutralizing epitopes in the porcine reproductive and respiratory syndrome virus

GP5 ectodomain. Journal of virology 76, 4241–4250.

Otsuka, S., Listowsky, I., Niitsu, Y., Urushizaki, I., 1980. Assembly of intra- and interspecies hybrid

apoferritins. The Journal of biological chemistry 255, 6234–6237.

Patel, K.G., Swartz, J.R., 2011. Surface functionalization of virus-like particles by direct conjugation using

azide-alkyne click chemistry. Bioconjugate chemistry 22, 376–387.

https://doi.org/10.1021/bc100367u

Pereira, E.O., Kolotilin, I., Conley, A.J., Menassa, R., 2014. Production and characterization of in planta

transiently produced polygalacturanase from Aspergillus niger and its fusions with hydrophobin or

ELP tags . BMC biotechnology . https://doi.org/10.1186/1472-6750-14-59

Piras, F., Bollard, S., Laval, F., Joisel, F., Reynaud, G., Charreyre, C., Andreoni, C., Juillard, V., 2005.

Porcine reproductive and respiratory syndrome (PRRS) virus-specific interferon-gamma(+) T-cell

responses after PRRS virus infection or vaccination with an inactivated PRRS vaccine. Viral

immunology 18, 381–389. https://doi.org/10.1089/vim.2005.18.381

Plagemann, P.G.W., Rowland, R.R.R., Faaberg, K.S., 2002. The primary neutralization epitope of porcine

respiratory and reproductive syndrome virus strain VR-2332 is located in the middle of the GP5

ectodomain. Archives of virology 147, 2327–2347. https://doi.org/10.1007/s00705-002-0887-2

Plotkin, S., 2014. History of vaccination. Proceedings of the National Academy of Sciences of the United

States of America 111, 12283–12287. https://doi.org/10.1073/pnas.1400472111

Plummer, E.M., Manchester, M., 2011. Viral nanoparticles and virus‐like particles: platforms for

contemporary vaccine design . Wiley Interdisciplinary Reviews: Nanomedicine and

Nanobiotechnology . https://doi.org/10.1002/wnan.119

Quinlan, R.A., Zhang, Y., Lansbury, A., Williamson, I., Pohl, E., Sun, F., 2013. Changes in the quaternary

structure and function of MjHSP16.5 attributable to deletion of the IXI motif and introduction of the

substitution, R107G, in the alpha-crystallin domain. Philosophical transactions of the Royal Society

of London. Series B, Biological sciences 368, 20120327. https://doi.org/10.1098/rstb.2012.0327

Renukaradhya, G.J., Dwivedi, V., Manickam, C., Binjawadagi, B., Benfield, D., 2012. Mucosal vaccines to

69

prevent porcine reproductive and respiratory syndrome: a new perspective . Animal health research

reviews / Conference of Research Workers in Animal Diseases .

https://doi.org/10.1017/S1466252312000023

Renukaradhya, G.J., Meng, X.-J., Calvert, J.G., Roof, M., Lager, K.M., 2015. Inactivated and subunit

vaccines against porcine reproductive and respiratory syndrome: Current status and future direction .

Vaccine . https://doi.org/10.1016/j.vaccine.2015.04.102

Richmond, J.E., Cohen, B.J., 1982. Electron microscopy of hepatitis B core antigen synthesized in E. coli .

Nature . https://doi.org/10.1038/296677a0

Riedel, S., 2005. Edward Jenner and the history of smallpox and vaccination . Proceedings (Baylor

University. Medical Center) .

Rodríguez-Limas, W.A., Sekar, K., Tyo, K.E.J., 2013. Virus-like particles: the future of microbial factories

and cell-free systems as platforms for vaccine development . Current opinion in biotechnology .

https://doi.org/10.1016/j.copbio.2013.02.008

Roldão, A., Mellado, M.C.M., Castilho, L.R., Carrondo, M.J.T., Alves, P.M., 2010. Virus-like particles in

vaccine development . Expert review of vaccines . https://doi.org/10.1586/erv.10.115

Rosas, G., Fragoso, G., Ainciart, N., Esquivel-Guadarrama, F., Santana, A., Bobes, R.J., Ramirez-Pliego,

O., Toledo, A., Cruz-Revilla, C., Meneses, G., Berguer, P., Goldbaum, F.A., Sciutto, E., 2006.

Brucella spp. lumazine synthase: a novel adjuvant and antigen delivery system to effectively induce

oral immunity. Microbes and infection 8, 1277–1286. https://doi.org/10.1016/j.micinf.2005.12.006

Rossi, A.H., Farias, A., Fernandez, J.E., Bonomi, H.R., Goldbaum, F.A., Berguer, P.M., 2015. Brucella

spp. Lumazine Synthase Induces a TLR4-Mediated Protective Response against B16 Melanoma in

Mice. PloS one 10, e0126827. https://doi.org/10.1371/journal.pone.0126827

Saunders, K., Sainsbury, F., Lomonossoff, G.P., 2009. Efficient generation of cowpea mosaic virus empty

virus-like particles by the proteolytic processing of precursors in insect cells and plants. Virology

393, 329–337. https://doi.org/10.1016/j.virol.2009.08.023

Schneider, Thomas D, Schneider, T D, Stephens, R M, Stephens, R Michael, 1990. Sequence logos: A new

way to display consensus sequences . Nucleic Acids Research .

https://doi.org/10.1093/nar/18.20.6097

Smith, C.A., 1988. Estimation of sedimentation coefficients and frictional ratios of globular proteins.

Biochemical Education 16, 104–106. https://doi.org/https://doi.org/10.1016/0307-4412(88)90075-1

70

Snijder, E.J., Dobbe, J.C., Spaan, W.J.M., 2003. Heterodimerization of the two major envelope proteins is

essential for arterivirus infectivity. Journal of virology 77, 97–104.

https://doi.org/10.1128/jvi.77.1.97-104.2003

Snijder, E.J., Kikkert, M., Fang, Y., 2013. Arterivirus molecular biology and pathogenesis. The Journal of

general virology 94, 2141–2163. https://doi.org/10.1099/vir.0.056341-0

Sorensen, B.R., Eppel, J.T., Shea, M.A., 2001. Paramecium calmodulin mutants defective in ion channel

regulation associate with melittin in the absence of calcium but require it for tertiary collapse.

Biochemistry 40, 896–903. https://doi.org/10.1021/bi0023091

Sun, S., Rao, S., Holdaway, H.A., Akahata, W., Nabel, G.J., Rossmann, M.G., Lewis, M.G., Higgs, S.,

Yang, Z.-Y., Andersen, H., Kong, W.-P., 2010. A virus-like particle vaccine for epidemic

Chikungunya virus protects nonhuman primates against infection . Nature Medicine .

https://doi.org/10.1038/nm.2105

Thuenemann, E.C., Meyers, A.E., Verwey, J., Rybicki, E.P., Lomonossoff, G.P., 2013. A method for rapid

production of heteromultimeric protein complexes in plants: assembly of protective bluetongue virus‐

like particles . Plant Biotechnology Journal . https://doi.org/10.1111/pbi.12076

Truffi, M., Fiandra, L., Sorrentino, L., Monieri, M., Corsi, F., Mazzucchelli, S., 2016. Ferritin nanocages:

A biological platform for drug delivery, imaging and theranostics in cancer. Pharmacological

Research. https://doi.org/10.1016/j.phrs.2016.03.002

Uchida, M., Klem, M.T., Allen, M., Suci, P., Flenniken, M., Gillitzer, E., Varpness, Z., Liepold, L.O.,

Young, M., Douglas, T., 2007. Biological Containers: Protein Cages as Multifunctional

Nanoplatforms . Advanced Materials . https://doi.org/10.1002/adma.200601168

Unzai, S., 2018. Analytical ultracentrifugation in structural biology. Biophysical reviews 10, 229–233.

https://doi.org/10.1007/s12551-017-0340-0

Velikovsky, C.A., Cassataro, J., Giambartolomei, G.H., Goldbaum, F.A., Estein, S., Bowden, R.A., Bruno,

L., Fossati, C.A., Spitz, M., 2002. A DNA vaccine encoding lumazine synthase from Brucella abortus

induces protective immunity in BALB/c mice. Infection and immunity 70, 2507–2511.

https://doi.org/10.1128/iai.70.5.2507-2511.2002

Velikovsky, C.A., Goldbaum, F.A., Cassataro, J., Estein, S., Bowden, R.A., Bruno, L., Fossati, C.A.,

Giambartolomei, G.H., 2003. Brucella lumazine synthase elicits a mixed Th1-Th2 immune response

and reduces infection in mice challenged with Brucella abortus 544 independently of the adjuvant

formulation used. Infection and immunity 71, 5750–5755. https://doi.org/10.1128/iai.71.10.5750-

71

5755.2003

Verheije, M.H., Welting, T.J.M., Jansen, H.T., Rottier, P.J.M., Meulenberg, J.J.M., 2002. Chimeric

arteriviruses generated by swapping of the M protein ectodomain rule out a role of this domain in

viral targeting. Virology 303, 364–373. https://doi.org/10.1006/viro.2002.1711

Vu, H.L.X., Kwon, B., Yoon, K.-J., Laegreid, W.W., Pattnaik, A.K., Osorio, F.A., 2011. Immune evasion

of porcine reproductive and respiratory syndrome virus through glycan shielding involves both

glycoprotein 5 as well as glycoprotein 3. Journal of virology 85, 5555–5564.

https://doi.org/10.1128/JVI.00189-11

Wang, S., Al-Soodani, A.T., Thomas, G.C., Buck-Koehntop, B.A., Woycechowsky, K.J., 2018. A Protein-

Capsid-Based System for Cell Delivery of Selenocysteine. Bioconjugate chemistry 29, 2332–2342.

https://doi.org/10.1021/acs.bioconjchem.8b00302

Wensvoort, G., Terpstra, C., Pol, J.M., ter Laak, E.A., Bloemraad, M., de Kluyver, E.P., Kragten, C., van

Buiten, L., den Besten, A., Wagenaar, F., 1991. Mystery swine disease in The Netherlands: the

isolation of Lelystad virus. The Veterinary quarterly 13, 121–130.

https://doi.org/10.1080/01652176.1991.9694296

Wissink, E.H.J., Kroese, M. V, van Wijk, H.A.R., Rijsewijk, F.A.M., Meulenberg, J.J.M., Rottier, P.J.M.,

2005. Envelope protein requirements for the assembly of infectious virions of porcine reproductive

and respiratory syndrome virus. Journal of virology 79, 12495–12506.

https://doi.org/10.1128/JVI.79.19.12495-12506.2005

Wissink, E.H.J., van Wijk, H.A.R., Kroese, M. V, Weiland, E., Meulenberg, J.J.M., Rottier, P.J.M., van

Rijn, P.A., 2003. The major envelope protein, GP5, of a European porcine reproductive and

respiratory syndrome virus contains a neutralization epitope in its N-terminal ectodomain. The

Journal of general virology 84, 1535–1543. https://doi.org/10.1099/vir.0.18957-0

Worsdorfer, B., Pianowski, Z., Hilvert, D., 2012. Efficient in vitro encapsulation of protein cargo by an

engineered protein container. Journal of the American Chemical Society 134, 909–911.

https://doi.org/10.1021/ja211011k

Worsdorfer, B., Woycechowsky, K.J., Hilvert, D., 2011. Directed evolution of a protein container. Science

(New York, N.Y.) 331, 589–592. https://doi.org/10.1126/science.1199081

Yoshimura, H., Edwards, E., Uchida, M., McCoy, K., Roychoudhury, R., Schwarz, B., Patterson, D.,

Douglas, T., 2016. Two-Dimensional Crystallization of P22 Virus-Like Particles. The journal of

physical chemistry. B 120, 5938–5944. https://doi.org/10.1021/acs.jpcb.6b01425

72

Yu, M., Qiu, Y., Chen, J., Jiang, W., 2016. Enhanced humoral and cellular immune responses to PRRS

virus GP5 glycoprotein by DNA prime-adenovirus boost vaccination in mice. Virus genes 52, 228–

234. https://doi.org/10.1007/s11262-016-1293-2

Zhang, X., Meining, W., Fischer, M., Bacher, A., Ladenstein, R., 2001. X-ray structure analysis and

crystallographic refinement of lumazine synthase from the hyperthermophile Aquifex aeolicus at 1.6

Å resolution: determinants of thermostability revealed from structural comparisons1 1Edited by R.

Huber. Journal of Molecular Biology 306, 1099–1114.

https://doi.org/https://doi.org/10.1006/jmbi.2000.4435

Zylberman, V., Craig, P.O., Klinke, S., Braden, B.C., Cauerhff, A., Goldbaum, F.A., 2004. High Order

Quaternary Arrangement Confers Increased Structural Stability to Brucella sp. Lumazine Synthase.

Journal of Biological Chemistry 279, 8093–8101. https://doi.org/10.1074/jbc.M312035200

73

Appendices

Appendix 1. Amino acid sequences of all proteins produced recombinantly.

Brucella lumazine synthase (BLS):

MGSSHHHHHHENLYFQSNAKTSFKIAFIQARWHADIVDEARKSFVAELAAKTGGSVEVEIFD

VPGAYEIPLHAKTLARTGRYAAIVGAAFVIDGGIYRHDFVATAVINGMMQVQLETEVPVLSV

VLTPHHFHESKEHHDFFHA HFKVKGVEAAHAALQIVSERSRIAALV

GP5-BLS:

MGSSHHHHHHENLYFQSNANASNDSSSHLQLIYNLTLCELNGTDGGSGGSGGSGGSKTSFKI

AFIQARWHADIVDEARKSFVAELAAKTGGSVEVEIFDVPGAYEIPLHAKTLARTGRYAAIVG

AAFVIDGGIYRHDFVATAVINGMMQVQLETEVPVLSVVLTPHHFHESKEHHDFFHAHFKVK

GVEAAHAALQIVSERSRIAALV

GP5-M-BLS:

MGSSHHHHHHENLYFQSNANASNDSSSHLQLIYNLTLCELNGTDGGSGGSGGSGGS

MGSSLDDFCHDSTAPQKVGGSGGSGGSGGSKTSFKIAFIQARWHADIVDEARKSFVAELAAK

TGGSVEVEIFDVPGAYEIPLHAKTLARTGRYAAIVGAAFVIDGGIYRHDFVATAVINGMMQV

QLETEVPVLSVVLTPHHFHESKEHHDFFHAHFKVKGVEAAHAALQIVSERSRIAALV

M-GP5-BLS:

MGSSHHHHHHENLYFQSNAHMGSSLDDFCHDSTAPQKVGGSGGSGGSGGSNASNDSSSHLQ

LIYNLTLCELNGTDGGSGGSGGSGGSKTSFKIAFIQARWHADIVDEARKSFVAELAAKTGGS

VEVEIFDVPGAYEIPLHAKTLARTGRYAAIVGAAFVIDGGIYRHDFVATAVINGMMQVQLET

EVPVLSVVLTPHHFHESKEHHDFFHAHFKVKGVEAAHAALQIVSERSRIAALV

Aquifex aeolicus lumazine synthase (AaLS):

MGSSHHHHHHSSGLVPRGSHMQIYEGKLTAEGLRFGIVASRFNHALVDRLVEGAIDCIVRHG

GREEDITLVRVPGSWEIPVAAGELARKEDIDAVIAIGVLIRGATPHFDYIASEVSKGLANLSLE

LRKPITFGVITADTLEQAIER AGTKHGNKGWEAALSAIEMANLFKSLRLE

74

AaLS-GP5:

MQIYEGKLTAEGLRFGIVASRFNHALVDRLVEGAIDCIVRHGGREEDITLVRVPGSWEIPVAA

GELARKEDIDAVIAIGVLIRGATPHFDYIASEVSKGLANLSLELRKPITFGVITADTLEQAIERA

GTKHGNKGWEAALSAIEMANLFKSLRGGSGGSGGSGGSNASNDSSSHLQLIYNLTLCELNGT

DLEHHHHHH

Methanococcus jannaschii small heat shock protein (sHSP-16.5):

MFGRDPFDSLFERMFKEFFATPMTGTTMIQSSTGIQISGKGFMPISIIEGDQHIKVIAWLPGVN

KEDIILNAVGDTLEIRAKRSPLMITESERIIYSEIPEEEEIYRTIKLPATVKEENASAKFENGVLS

VILPKAESSIKKGINIELEHHHHHH

sHSP-16.5-GP5:

MFGRDPFDSLFERMFKEFFATPMTGTTMIQSSTGIQISGKGFMPISIIEGDQHIKVIAWLPGVN

KEDIILNAVGDTLEIRAKRSPLMITESERIIYSEIPEEEEIYRTIKLPATVKEENASAKFENGVLS

VILPKAESSIKKGINIEGGSGGSGGSGGSNASNDSSSHLQLIYNLTLCELNGTDLEHHHHHH

Helicobacter pylori ferritin (ferritin):

MGSSHHHHHHSSGLVPRGSHMLSKDIIKLLNEQVNKEMNSSNLYMSMSSWCYTHSLDGAGL

FLFDHAAEEYEHAKKLIVFLNENNVPVQLTSISAPEHKFEGLTQIFQKAYGHEQHISESINNIV

DHAIKSKDHATFNFLQWYVAEQHEEEVLFKDILDKIELIGNENHGLYLADQYVKGIAKSRKS

LE

GP5-ferritin:

MGSSHHHHHHSSGLVPRGSHNASNDSSSHLQLIYNLTLCELNGTDGGSGGSGGSGGSMLSK

DIIKLLNEQVNKEMNSSNLYMSMSSWCYTHSLDGAGLFLFDHAAEEYEHAKKLIVFLNENN

VPVQLTSISAPEHKFEGLTQIFQKAYGHEQHISESINNIVDHAIKSKDHATFNFLQWYVAEQHE

EEVLFKDILDKIELIGNENHGLYLADQYVKGIAKSRKS

75

Curriculum Vitae

Name: Ondre Harper

Post-secondary Education and Degrees:

Master of Science in Biology 2017- present

Western University, London, Ontario

Thesis title: “Engineering self-assembling proteins to produce

a safe and effective vaccine for Porcine Reproductive

and Respiratory Syndrome”

Bachelor of Medical Sciences Honors (4 Year) 2017

Western University, London, Ontario

Honors Specialization in Biochemistry

Major in Microbiology and Immunology

Honors thesis title: “Investigation of the dormant IS200 transposon

conserved in Salmonella”

Honors and Awards:

Western University Continuing Admission Scholarship 2013-2016

Related Work Experience:

Teaching Assistant, Methods in Biology (Bio 2290), Western University 2017-2019

Poster Presentations:

• Harper O., Menassa R., Garnham C. (2018, July). Engineering self-assembling proteins

to produce a safe and effective vaccine for Porcine Reproductive and Respiratory

Syndrome. Poster presented at the Synthetic Biology Symposium 3.0, London, Ontario.

• Harper O., Menassa R., Garnham C. (2019, May). Engineering self-assembling proteins

to produce a safe and effective vaccine for Porcine Reproductive and Respiratory

Syndrome. Poster presented at the Synthetic Biology Symposium 4.0, Waterloo, Ontario.